Token |
|
Token ID | 198 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
LLOW | +0.469 |
tten | +0.453 |
�� | +0.442 |
Swordsman | +0.437 |
FACE | +0.430 |
Neg logit contributions | |
ippers | -0.433 |
ipper | -0.423 |
rones | -0.337 |
lems | -0.336 |
ploy | -0.329 |
Token |
|
Token ID | 198 |
Feature activation | +9.07e-01 |
Pos logit contributions | |
LLOW | +1.097 |
tten | +1.062 |
�� | +1.015 |
Swordsman | +1.007 |
FACE | +1.006 |
Neg logit contributions | |
ippers | -1.025 |
ipper | -0.998 |
rones | -0.796 |
lems | -0.788 |
lihood | -0.788 |
Token | 12 |
Token ID | 1065 |
Feature activation | +2.70e-01 |
Pos logit contributions | |
LLOW | +0.581 |
tten | +0.561 |
�� | +0.545 |
FACE | +0.525 |
Swordsman | +0.521 |
Neg logit contributions | |
ippers | -0.788 |
ipper | -0.772 |
rones | -0.640 |
lihood | -0.637 |
lems | -0.637 |
Token | . |
Token ID | 13 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
LLOW | +0.198 |
tten | +0.192 |
�� | +0.186 |
Swordsman | +0.182 |
FACE | +0.180 |
Neg logit contributions | |
ippers | -0.210 |
ipper | -0.205 |
rones | -0.166 |
lems | -0.165 |
Carney | -0.163 |
Token | Donkey |
Token ID | 43823 |
Feature activation | +5.10e+00 |
Pos logit contributions | |
ippers | +0.293 |
ipper | +0.287 |
rones | +0.233 |
lems | +0.233 |
Carney | +0.230 |
Neg logit contributions | |
LLOW | -0.266 |
tten | -0.259 |
�� | -0.253 |
Swordsman | -0.245 |
FACE | -0.241 |
Token | Kong |
Token ID | 9071 |
Feature activation | +8.80e+00 |
Pos logit contributions | |
LLOW | +1.295 |
tten | +1.110 |
�� | +1.058 |
FACE | +1.042 |
Swordsman | +1.016 |
Neg logit contributions | |
ippers | -6.266 |
ipper | -6.148 |
lems | -5.563 |
lihood | -5.526 |
rones | -5.511 |
Token | Country |
Token ID | 12946 |
Feature activation | +4.44e+00 |
Pos logit contributions | |
LLOW | +4.867 |
tten | +4.641 |
Swordsman | +4.473 |
ADRA | +4.230 |
�� | +4.219 |
Neg logit contributions | |
ippers | -7.830 |
ipper | -7.657 |
lems | -6.602 |
rones | -6.588 |
lihood | -6.500 |
Token | 2 |
Token ID | 362 |
Feature activation | +3.01e+00 |
Pos logit contributions | |
LLOW | +3.068 |
tten | +2.974 |
�� | +2.811 |
Swordsman | +2.786 |
Bleach | +2.699 |
Neg logit contributions | |
ippers | -3.570 |
ipper | -3.505 |
lems | -2.896 |
rones | -2.876 |
lihood | -2.864 |
Token | : |
Token ID | 25 |
Feature activation | +3.11e+00 |
Pos logit contributions | |
LLOW | +2.378 |
tten | +2.320 |
�� | +2.248 |
Swordsman | +2.189 |
FACE | +2.170 |
Neg logit contributions | |
ippers | -2.161 |
ipper | -2.109 |
rones | -1.676 |
lems | -1.674 |
Carney | -1.653 |
Token | D |
Token ID | 360 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
LLOW | +2.147 |
tten | +2.057 |
Swordsman | +1.983 |
�� | +1.972 |
FACE | +1.930 |
Neg logit contributions | |
ippers | -2.475 |
ipper | -2.441 |
rones | -2.034 |
mast | -1.998 |
ploy | -1.998 |
Token | iddy |
Token ID | 34208 |
Feature activation | +4.18e+00 |
Pos logit contributions | |
LLOW | +0.875 |
tten | +0.853 |
�� | +0.831 |
Swordsman | +0.801 |
ADRA | +0.797 |
Neg logit contributions | |
ippers | -0.922 |
ipper | -0.902 |
lems | -0.733 |
rones | -0.728 |
Carney | -0.727 |
Token | in |
Token ID | 287 |
Feature activation | -3.26e-01 |
Pos logit contributions | |
ippers | +0.234 |
ipper | +0.228 |
rones | +0.187 |
lems | +0.185 |
Carney | +0.184 |
Neg logit contributions | |
LLOW | -0.213 |
tten | -0.208 |
�� | -0.200 |
Swordsman | -0.197 |
FACE | -0.193 |
Token | Japan |
Token ID | 2869 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ippers | +0.237 |
ipper | +0.231 |
rones | +0.185 |
lems | +0.184 |
Carney | +0.182 |
Neg logit contributions | |
LLOW | -0.256 |
tten | -0.249 |
�� | -0.244 |
Swordsman | -0.237 |
FACE | -0.233 |
Token | if |
Token ID | 611 |
Feature activation | +3.14e-02 |
Pos logit contributions | |
LLOW | +1.266 |
tten | +1.232 |
�� | +1.184 |
Swordsman | +1.163 |
FACE | +1.141 |
Neg logit contributions | |
ippers | -1.441 |
ipper | -1.410 |
rones | -1.153 |
lems | -1.145 |
lihood | -1.137 |
Token | everyone |
Token ID | 2506 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
LLOW | +0.028 |
tten | +0.027 |
�� | +0.027 |
Swordsman | +0.026 |
FACE | +0.026 |
Neg logit contributions | |
ippers | -0.019 |
ipper | -0.019 |
rones | -0.014 |
lems | -0.014 |
Carney | -0.014 |
Token | comes |
Token ID | 2058 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
LLOW | +1.373 |
tten | +1.337 |
�� | +1.292 |
Swordsman | +1.268 |
FACE | +1.239 |
Neg logit contributions | |
ippers | -1.271 |
ipper | -1.249 |
rones | -0.991 |
lems | -0.984 |
ploy | -0.979 |
Token | out |
Token ID | 503 |
Feature activation | +7.75e+00 |
Pos logit contributions | |
LLOW | +1.789 |
tten | +1.714 |
�� | +1.652 |
Swordsman | +1.622 |
FACE | +1.593 |
Neg logit contributions | |
ippers | -2.083 |
ipper | -2.041 |
rones | -1.682 |
lems | -1.656 |
mast | -1.653 |
Token | and |
Token ID | 290 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
LLOW | +4.973 |
tten | +4.835 |
Swordsman | +4.540 |
Bleach | +4.511 |
�� | +4.449 |
Neg logit contributions | |
ippers | -6.126 |
ipper | -5.951 |
rones | -5.034 |
ploy | -5.004 |
lems | -4.940 |
Token | has |
Token ID | 468 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
LLOW | +1.433 |
tten | +1.401 |
�� | +1.364 |
Swordsman | +1.338 |
FACE | +1.323 |
Neg logit contributions | |
ippers | -0.955 |
ipper | -0.929 |
rones | -0.701 |
lems | -0.694 |
ploy | -0.689 |
Token | a |
Token ID | 257 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
LLOW | +1.117 |
tten | +1.096 |
�� | +1.058 |
Swordsman | +1.032 |
FACE | +1.025 |
Neg logit contributions | |
ippers | -1.013 |
ipper | -0.995 |
rones | -0.789 |
lems | -0.782 |
ploy | -0.774 |
Token | great |
Token ID | 1049 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
LLOW | +0.857 |
tten | +0.838 |
�� | +0.815 |
Swordsman | +0.792 |
FACE | +0.783 |
Neg logit contributions | |
ippers | -0.862 |
ipper | -0.842 |
lems | -0.678 |
rones | -0.677 |
Carney | -0.668 |
Token | time |
Token ID | 640 |
Feature activation | -1.93e-03 |
Pos logit contributions | |
ippers | +0.800 |
ipper | +0.783 |
rones | +0.630 |
lems | +0.627 |
Carney | +0.623 |
Neg logit contributions | |
LLOW | -0.787 |
tten | -0.767 |
�� | -0.752 |
Swordsman | -0.725 |
ADRA | -0.719 |
Token | p |
Token ID | 279 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ippers | +0.212 |
ipper | +0.207 |
rones | +0.161 |
lems | +0.160 |
Carney | +0.159 |
Neg logit contributions | |
LLOW | -0.270 |
tten | -0.264 |
�� | -0.259 |
Swordsman | -0.251 |
ADRA | -0.249 |
Token | . |
Token ID | 13 |
Feature activation | -1.34e+00 |
Pos logit contributions | |
ippers | +1.129 |
ipper | +1.105 |
rones | +0.882 |
lems | +0.881 |
Carney | +0.872 |
Neg logit contributions | |
LLOW | -1.187 |
tten | -1.162 |
�� | -1.148 |
ADRA | -1.102 |
Swordsman | -1.099 |
Token | m |
Token ID | 76 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
ippers | +0.894 |
ipper | +0.871 |
rones | +0.677 |
lems | +0.674 |
Carney | +0.667 |
Neg logit contributions | |
LLOW | -1.134 |
tten | -1.108 |
�� | -1.089 |
Swordsman | -1.055 |
ADRA | -1.046 |
Token | .: |
Token ID | 11207 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
LLOW | +0.444 |
tten | +0.432 |
�� | +0.420 |
Swordsman | +0.403 |
ADRA | +0.398 |
Neg logit contributions | |
ippers | -0.638 |
ipper | -0.626 |
rones | -0.523 |
lems | -0.522 |
Carney | -0.518 |
Token | The |
Token ID | 383 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
LLOW | +1.318 |
tten | +1.254 |
�� | +1.229 |
Swordsman | +1.210 |
FACE | +1.194 |
Neg logit contributions | |
ippers | -1.409 |
ipper | -1.378 |
rones | -1.112 |
lems | -1.106 |
lihood | -1.090 |
Token | number |
Token ID | 1271 |
Feature activation | +7.55e+00 |
Pos logit contributions | |
LLOW | +1.421 |
tten | +1.371 |
�� | +1.351 |
Swordsman | +1.306 |
FACE | +1.287 |
Neg logit contributions | |
ippers | -1.374 |
ipper | -1.339 |
rones | -1.074 |
lems | -1.066 |
lihood | -1.057 |
Token | of |
Token ID | 286 |
Feature activation | +2.78e+00 |
Pos logit contributions | |
LLOW | +5.265 |
tten | +5.057 |
�� | +4.990 |
FACE | +4.764 |
Swordsman | +4.730 |
Neg logit contributions | |
ippers | -5.884 |
ipper | -5.780 |
lems | -4.735 |
rones | -4.672 |
mast | -4.612 |
Token | injured |
Token ID | 6686 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
LLOW | +1.873 |
tten | +1.827 |
�� | +1.782 |
Swordsman | +1.720 |
FACE | +1.699 |
Neg logit contributions | |
ippers | -2.290 |
ipper | -2.252 |
rones | -1.843 |
lems | -1.830 |
mast | -1.813 |
Token | has |
Token ID | 468 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
LLOW | +1.337 |
tten | +1.299 |
�� | +1.266 |
Swordsman | +1.243 |
FACE | +1.225 |
Neg logit contributions | |
ippers | -1.163 |
ipper | -1.144 |
lems | -0.896 |
rones | -0.885 |
mast | -0.876 |
Token | reached |
Token ID | 4251 |
Feature activation | +8.75e-01 |
Pos logit contributions | |
LLOW | +1.076 |
tten | +1.057 |
�� | +1.030 |
Swordsman | +0.983 |
FACE | +0.976 |
Neg logit contributions | |
ippers | -1.278 |
ipper | -1.255 |
lems | -1.022 |
rones | -1.022 |
lihood | -1.009 |
Token | 132 |
Token ID | 21761 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
LLOW | +0.634 |
tten | +0.619 |
�� | +0.603 |
Swordsman | +0.584 |
FACE | +0.575 |
Neg logit contributions | |
ippers | -0.686 |
ipper | -0.672 |
rones | -0.546 |
lems | -0.541 |
lihood | -0.536 |
Token | 11 |
Token ID | 1157 |
Feature activation | +3.40e-01 |
Pos logit contributions | |
LLOW | +0.435 |
tten | +0.424 |
�� | +0.415 |
Swordsman | +0.401 |
FACE | +0.398 |
Neg logit contributions | |
ippers | -0.410 |
ipper | -0.401 |
rones | -0.318 |
lems | -0.318 |
lihood | -0.315 |
Token | . |
Token ID | 13 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
LLOW | +0.250 |
tten | +0.242 |
�� | +0.235 |
Swordsman | +0.229 |
FACE | +0.227 |
Neg logit contributions | |
ippers | -0.265 |
ipper | -0.258 |
rones | -0.209 |
lems | -0.208 |
Carney | -0.206 |
Token | The |
Token ID | 383 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
ippers | +0.983 |
ipper | +0.963 |
rones | +0.787 |
lems | +0.785 |
Carney | +0.782 |
Neg logit contributions | |
LLOW | -0.847 |
tten | -0.826 |
�� | -0.810 |
Swordsman | -0.773 |
ADRA | -0.770 |
Token | Legend |
Token ID | 9883 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
LLOW | +0.367 |
tten | +0.355 |
�� | +0.345 |
Swordsman | +0.338 |
FACE | +0.329 |
Neg logit contributions | |
ippers | -0.441 |
ipper | -0.431 |
rones | -0.356 |
lems | -0.353 |
lihood | -0.350 |
Token | of |
Token ID | 286 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
LLOW | +1.775 |
tten | +1.732 |
�� | +1.697 |
Swordsman | +1.643 |
ADRA | +1.624 |
Neg logit contributions | |
ippers | -1.627 |
ipper | -1.588 |
rones | -1.263 |
lems | -1.258 |
Carney | -1.246 |
Token | Zelda |
Token ID | 22166 |
Feature activation | +7.41e+00 |
Pos logit contributions | |
LLOW | +0.747 |
tten | +0.724 |
�� | +0.705 |
Swordsman | +0.680 |
ADRA | +0.664 |
Neg logit contributions | |
ippers | -1.102 |
ipper | -1.082 |
rones | -0.904 |
lems | -0.903 |
Carney | -0.894 |
Token | : |
Token ID | 25 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
LLOW | +4.739 |
tten | +4.642 |
Swordsman | +4.330 |
�� | +4.168 |
FACE | +4.097 |
Neg logit contributions | |
ippers | -6.188 |
ipper | -5.992 |
rones | -5.161 |
lems | -5.075 |
ploy | -5.028 |
Token | O |
Token ID | 440 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
LLOW | +1.358 |
tten | +1.322 |
�� | +1.284 |
Swordsman | +1.254 |
FACE | +1.229 |
Neg logit contributions | |
ippers | -1.317 |
ipper | -1.286 |
rones | -1.042 |
lems | -1.028 |
Carney | -1.018 |
Token | car |
Token ID | 7718 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
LLOW | +0.781 |
tten | +0.769 |
�� | +0.748 |
Swordsman | +0.729 |
ADRA | +0.728 |
Neg logit contributions | |
ippers | -0.410 |
ipper | -0.395 |
rones | -0.285 |
lems | -0.281 |
lihood | -0.280 |
Token | ina |
Token ID | 1437 |
Feature activation | +6.65e-01 |
Pos logit contributions | |
tten | +1.007 |
LLOW | +1.003 |
�� | +0.935 |
Swordsman | +0.920 |
FACE | +0.915 |
Neg logit contributions | |
ippers | -1.233 |
ipper | -1.186 |
lihood | -1.012 |
lems | -1.009 |
ploy | -1.000 |
Token | of |
Token ID | 286 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
LLOW | +0.520 |
tten | +0.508 |
�� | +0.497 |
Swordsman | +0.482 |
FACE | +0.476 |
Neg logit contributions | |
ippers | -0.484 |
ipper | -0.473 |
rones | -0.377 |
lems | -0.375 |
Carney | -0.371 |
Token | K |
Token ID | 42 |
Feature activation | -9.44e-01 |
Pos logit contributions | |
ippers | +1.376 |
ipper | +1.353 |
rones | +1.163 |
lems | +1.159 |
Carney | +1.152 |
Neg logit contributions | |
LLOW | -0.600 |
tten | -0.575 |
�� | -0.555 |
Swordsman | -0.525 |
ADRA | -0.513 |
Token | people |
Token ID | 661 |
Feature activation | -4.78e-01 |
Pos logit contributions | |
ippers | +0.662 |
ipper | +0.646 |
rones | +0.511 |
lems | +0.509 |
Carney | +0.504 |
Neg logit contributions | |
LLOW | -0.763 |
tten | -0.745 |
�� | -0.729 |
Swordsman | -0.708 |
FACE | -0.699 |
Token | are |
Token ID | 389 |
Feature activation | +1.40e-01 |
Pos logit contributions | |
ippers | +0.357 |
ipper | +0.348 |
rones | +0.281 |
lems | +0.279 |
Carney | +0.275 |
Neg logit contributions | |
LLOW | -0.365 |
tten | -0.356 |
�� | -0.346 |
Swordsman | -0.337 |
FACE | -0.333 |
Token | still |
Token ID | 991 |
Feature activation | +2.05e+00 |
Pos logit contributions | |
LLOW | +0.106 |
tten | +0.103 |
�� | +0.100 |
Swordsman | +0.097 |
FACE | +0.097 |
Neg logit contributions | |
ippers | -0.106 |
ipper | -0.104 |
rones | -0.084 |
lems | -0.083 |
lihood | -0.082 |
Token | coming |
Token ID | 2406 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
LLOW | +1.518 |
tten | +1.473 |
�� | +1.405 |
Swordsman | +1.389 |
FACE | +1.384 |
Neg logit contributions | |
ippers | -1.553 |
ipper | -1.523 |
rones | -1.252 |
lems | -1.221 |
lihood | -1.216 |
Token | out |
Token ID | 503 |
Feature activation | +7.39e+00 |
Pos logit contributions | |
LLOW | +1.194 |
tten | +1.153 |
�� | +1.115 |
Swordsman | +1.090 |
FACE | +1.077 |
Neg logit contributions | |
ippers | -1.267 |
ipper | -1.237 |
rones | -1.010 |
lems | -0.995 |
Carney | -0.992 |
Token | with |
Token ID | 351 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
LLOW | +4.934 |
tten | +4.837 |
Swordsman | +4.514 |
Bleach | +4.487 |
FACE | +4.485 |
Neg logit contributions | |
ippers | -5.804 |
ipper | -5.637 |
rones | -4.670 |
ploy | -4.613 |
lihood | -4.587 |
Token | shorts |
Token ID | 22078 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
LLOW | +1.056 |
tten | +1.028 |
�� | +0.998 |
Swordsman | +0.977 |
FACE | +0.961 |
Neg logit contributions | |
ippers | -1.031 |
ipper | -1.006 |
lems | -0.806 |
rones | -0.806 |
lihood | -0.798 |
Token | from |
Token ID | 422 |
Feature activation | -6.03e-01 |
Pos logit contributions | |
LLOW | +2.294 |
tten | +2.228 |
�� | +2.139 |
Swordsman | +2.107 |
FACE | +2.085 |
Neg logit contributions | |
ippers | -2.305 |
ipper | -2.250 |
lems | -1.822 |
rones | -1.817 |
lihood | -1.816 |
Token | this |
Token ID | 428 |
Feature activation | +2.25e-01 |
Pos logit contributions | |
ippers | +0.443 |
ipper | +0.432 |
rones | +0.346 |
lems | +0.345 |
lihood | +0.341 |
Neg logit contributions | |
LLOW | -0.467 |
tten | -0.457 |
�� | -0.445 |
Swordsman | -0.433 |
FACE | -0.427 |
Token | site |
Token ID | 2524 |
Feature activation | +3.41e+00 |
Pos logit contributions | |
LLOW | +0.167 |
tten | +0.163 |
�� | +0.159 |
Swordsman | +0.154 |
FACE | +0.152 |
Neg logit contributions | |
ippers | -0.172 |
ipper | -0.168 |
rones | -0.137 |
lems | -0.136 |
lihood | -0.134 |
Token | the |
Token ID | 262 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
LLOW | +1.842 |
tten | +1.798 |
�� | +1.762 |
Swordsman | +1.706 |
ADRA | +1.684 |
Neg logit contributions | |
ippers | -1.833 |
ipper | -1.789 |
rones | -1.440 |
lems | -1.435 |
lihood | -1.424 |
Token | first |
Token ID | 717 |
Feature activation | +1.73e+00 |
Pos logit contributions | |
LLOW | +1.311 |
tten | +1.271 |
�� | +1.241 |
Swordsman | +1.206 |
FACE | +1.180 |
Neg logit contributions | |
ippers | -1.486 |
ipper | -1.443 |
lems | -1.199 |
rones | -1.192 |
lihood | -1.171 |
Token | step |
Token ID | 2239 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
LLOW | +1.194 |
tten | +1.168 |
�� | +1.130 |
Swordsman | +1.092 |
FACE | +1.078 |
Neg logit contributions | |
ippers | -1.425 |
ipper | -1.384 |
rones | -1.146 |
lems | -1.144 |
ploy | -1.104 |
Token | in |
Token ID | 287 |
Feature activation | -1.83e-01 |
Pos logit contributions | |
ippers | +0.819 |
ipper | +0.798 |
rones | +0.620 |
lems | +0.618 |
Carney | +0.611 |
Neg logit contributions | |
LLOW | -1.035 |
tten | -1.011 |
�� | -0.992 |
Swordsman | -0.962 |
ADRA | -0.953 |
Token | a |
Token ID | 257 |
Feature activation | +1.59e+00 |
Pos logit contributions | |
ippers | +0.133 |
ipper | +0.130 |
rones | +0.104 |
lems | +0.103 |
Carney | +0.101 |
Neg logit contributions | |
LLOW | -0.144 |
tten | -0.141 |
�� | -0.137 |
Swordsman | -0.133 |
FACE | -0.131 |
Token | sweeping |
Token ID | 18404 |
Feature activation | +7.31e+00 |
Pos logit contributions | |
LLOW | +1.167 |
tten | +1.146 |
�� | +1.102 |
Swordsman | +1.076 |
FACE | +1.066 |
Neg logit contributions | |
ippers | -1.227 |
ipper | -1.185 |
lems | -0.968 |
rones | -0.961 |
mast | -0.940 |
Token | review |
Token ID | 2423 |
Feature activation | +1.55e+00 |
Pos logit contributions | |
tten | +5.144 |
LLOW | +5.108 |
�� | +4.869 |
Swordsman | +4.745 |
ATH | +4.725 |
Neg logit contributions | |
ippers | -5.612 |
ipper | -5.329 |
lems | -4.394 |
rones | -4.387 |
ploy | -4.269 |
Token | of |
Token ID | 286 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
LLOW | +1.192 |
tten | +1.170 |
�� | +1.136 |
Swordsman | +1.104 |
FACE | +1.091 |
Neg logit contributions | |
ippers | -1.138 |
ipper | -1.108 |
rones | -0.892 |
lems | -0.890 |
lihood | -0.872 |
Token | the |
Token ID | 262 |
Feature activation | +5.03e-01 |
Pos logit contributions | |
LLOW | +0.818 |
tten | +0.799 |
�� | +0.780 |
Swordsman | +0.753 |
FACE | +0.746 |
Neg logit contributions | |
ippers | -0.799 |
ipper | -0.780 |
rones | -0.627 |
lems | -0.625 |
lihood | -0.614 |
Token | tobacco |
Token ID | 11219 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
LLOW | +0.392 |
tten | +0.382 |
�� | +0.374 |
Swordsman | +0.363 |
FACE | +0.358 |
Neg logit contributions | |
ippers | -0.369 |
ipper | -0.359 |
rones | -0.287 |
lems | -0.286 |
lihood | -0.282 |
Token | industry |
Token ID | 2831 |
Feature activation | +7.83e-01 |
Pos logit contributions | |
ippers | +0.327 |
ipper | +0.318 |
rones | +0.254 |
lems | +0.254 |
Carney | +0.250 |
Neg logit contributions | |
LLOW | -0.350 |
tten | -0.342 |
�� | -0.334 |
Swordsman | -0.324 |
FACE | -0.320 |
Token | not |
Token ID | 407 |
Feature activation | +9.86e-01 |
Pos logit contributions | |
LLOW | +0.234 |
tten | +0.230 |
�� | +0.223 |
Swordsman | +0.216 |
FACE | +0.213 |
Neg logit contributions | |
ippers | -0.219 |
ipper | -0.214 |
rones | -0.171 |
lems | -0.170 |
ploy | -0.167 |
Token | heard |
Token ID | 2982 |
Feature activation | +7.84e-01 |
Pos logit contributions | |
LLOW | +0.774 |
tten | +0.761 |
�� | +0.737 |
Swordsman | +0.709 |
FACE | +0.704 |
Neg logit contributions | |
ippers | -0.714 |
ipper | -0.696 |
rones | -0.556 |
lems | -0.553 |
mast | -0.541 |
Token | of |
Token ID | 286 |
Feature activation | +1.96e+00 |
Pos logit contributions | |
LLOW | +0.607 |
tten | +0.593 |
�� | +0.580 |
Swordsman | +0.560 |
FACE | +0.554 |
Neg logit contributions | |
ippers | -0.578 |
ipper | -0.564 |
rones | -0.451 |
lems | -0.449 |
Carney | -0.443 |
Token | the |
Token ID | 262 |
Feature activation | +2.18e-01 |
Pos logit contributions | |
LLOW | +1.203 |
tten | +1.155 |
�� | +1.119 |
Swordsman | +1.080 |
FACE | +1.058 |
Neg logit contributions | |
ippers | -1.770 |
ipper | -1.734 |
rones | -1.449 |
lems | -1.442 |
mast | -1.427 |
Token | Republican |
Token ID | 3415 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
LLOW | +0.148 |
tten | +0.143 |
�� | +0.139 |
Swordsman | +0.135 |
FACE | +0.132 |
Neg logit contributions | |
ippers | -0.182 |
ipper | -0.178 |
rones | -0.147 |
lems | -0.146 |
lihood | -0.144 |
Token | strategist |
Token ID | 25651 |
Feature activation | +7.28e+00 |
Pos logit contributions | |
LLOW | +2.008 |
tten | +1.965 |
�� | +1.900 |
Swordsman | +1.890 |
ATH | +1.834 |
Neg logit contributions | |
ippers | -2.090 |
ipper | -2.042 |
lems | -1.679 |
rones | -1.660 |
lihood | -1.638 |
Token | despite |
Token ID | 3805 |
Feature activation | -4.48e-01 |
Pos logit contributions | |
LLOW | +3.906 |
tten | +3.591 |
�� | +3.500 |
Swordsman | +3.447 |
Bleach | +3.350 |
Neg logit contributions | |
ippers | -6.924 |
ipper | -6.712 |
rones | -5.824 |
lems | -5.790 |
lihood | -5.633 |
Token | his |
Token ID | 465 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
ippers | +0.340 |
ipper | +0.332 |
rones | +0.268 |
lems | +0.267 |
Carney | +0.263 |
Neg logit contributions | |
LLOW | -0.336 |
tten | -0.328 |
�� | -0.320 |
Swordsman | -0.310 |
FACE | -0.306 |
Token | having |
Token ID | 1719 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
LLOW | +0.858 |
tten | +0.839 |
�� | +0.812 |
Swordsman | +0.790 |
FACE | +0.784 |
Neg logit contributions | |
ippers | -0.848 |
ipper | -0.830 |
rones | -0.671 |
lems | -0.669 |
mast | -0.652 |
Token | been |
Token ID | 587 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
LLOW | +0.887 |
tten | +0.874 |
�� | +0.844 |
Swordsman | +0.817 |
FACE | +0.812 |
Neg logit contributions | |
ippers | -0.921 |
ipper | -0.902 |
rones | -0.732 |
lems | -0.726 |
mast | -0.716 |
Token | widely |
Token ID | 6768 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
ippers | +0.882 |
ipper | +0.861 |
rones | +0.692 |
lems | +0.688 |
lihood | +0.675 |
Neg logit contributions | |
LLOW | -0.926 |
tten | -0.903 |
�� | -0.881 |
Swordsman | -0.853 |
FACE | -0.847 |
Token | grapp |
Token ID | 21338 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
LLOW | +0.512 |
tten | +0.498 |
�� | +0.485 |
Swordsman | +0.477 |
FACE | +0.463 |
Neg logit contributions | |
ippers | -0.563 |
ipper | -0.548 |
rones | -0.447 |
lems | -0.445 |
lihood | -0.440 |
Token | ler |
Token ID | 1754 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
LLOW | +2.571 |
tten | +2.559 |
�� | +2.472 |
Swordsman | +2.460 |
FACE | +2.424 |
Neg logit contributions | |
ippers | -1.046 |
ipper | -0.988 |
lihood | -0.687 |
rones | -0.671 |
Carney | -0.661 |
Token | in |
Token ID | 287 |
Feature activation | -7.48e-01 |
Pos logit contributions | |
ippers | +0.551 |
ipper | +0.540 |
rones | +0.435 |
lems | +0.434 |
Carney | +0.431 |
Neg logit contributions | |
LLOW | -0.534 |
tten | -0.521 |
�� | -0.511 |
Swordsman | -0.490 |
ADRA | -0.488 |
Token | the |
Token ID | 262 |
Feature activation | +9.62e-01 |
Pos logit contributions | |
ippers | +0.564 |
ipper | +0.551 |
rones | +0.443 |
lems | +0.442 |
Carney | +0.437 |
Neg logit contributions | |
LLOW | -0.566 |
tten | -0.552 |
�� | -0.541 |
Swordsman | -0.522 |
ADRA | -0.516 |
Token | Street |
Token ID | 3530 |
Feature activation | +6.05e+00 |
Pos logit contributions | |
LLOW | +0.710 |
tten | +0.690 |
�� | +0.674 |
Swordsman | +0.658 |
FACE | +0.643 |
Neg logit contributions | |
ippers | -0.745 |
ipper | -0.728 |
rones | -0.591 |
lems | -0.586 |
lihood | -0.582 |
Token | Fighter |
Token ID | 15411 |
Feature activation | +7.20e+00 |
Pos logit contributions | |
LLOW | +3.952 |
tten | +3.810 |
�� | +3.684 |
Swordsman | +3.673 |
FACE | +3.563 |
Neg logit contributions | |
ippers | -5.074 |
ipper | -4.962 |
rones | -4.113 |
lems | -4.086 |
lihood | -4.021 |
Token | series |
Token ID | 2168 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
LLOW | +4.865 |
tten | +4.630 |
�� | +4.546 |
Swordsman | +4.493 |
Bleach | +4.324 |
Neg logit contributions | |
ippers | -5.866 |
ipper | -5.724 |
lems | -4.758 |
rones | -4.717 |
mast | -4.638 |
Token | to |
Token ID | 284 |
Feature activation | +5.12e-01 |
Pos logit contributions | |
LLOW | +1.821 |
tten | +1.772 |
�� | +1.733 |
Swordsman | +1.681 |
FACE | +1.650 |
Neg logit contributions | |
ippers | -2.070 |
ipper | -2.023 |
rones | -1.661 |
lems | -1.649 |
Carney | -1.629 |
Token | ever |
Token ID | 1683 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
LLOW | +0.407 |
tten | +0.400 |
�� | +0.387 |
Swordsman | +0.380 |
FACE | +0.374 |
Neg logit contributions | |
ippers | -0.364 |
ipper | -0.355 |
rones | -0.284 |
lems | -0.282 |
lihood | -0.278 |
Token | have |
Token ID | 423 |
Feature activation | -1.63e-01 |
Pos logit contributions | |
ippers | +0.888 |
ipper | +0.867 |
rones | +0.687 |
lems | +0.685 |
Carney | +0.679 |
Neg logit contributions | |
LLOW | -0.981 |
tten | -0.957 |
�� | -0.941 |
Swordsman | -0.907 |
ADRA | -0.901 |
Token | a |
Token ID | 257 |
Feature activation | +9.56e-01 |
Pos logit contributions | |
ippers | +0.118 |
ipper | +0.115 |
rones | +0.092 |
lems | +0.091 |
Carney | +0.090 |
Neg logit contributions | |
LLOW | -0.128 |
tten | -0.125 |
�� | -0.122 |
Swordsman | -0.119 |
FACE | -0.117 |
Token | poll |
Token ID | 3278 |
Feature activation | +2.84e+00 |
Pos logit contributions | |
LLOW | +0.946 |
tten | +0.917 |
�� | +0.902 |
Swordsman | +0.848 |
ATH | +0.828 |
Neg logit contributions | |
ippers | -1.566 |
ipper | -1.535 |
lems | -1.324 |
rones | -1.311 |
lihood | -1.283 |
Token | conducted |
Token ID | 5952 |
Feature activation | +4.24e-01 |
Pos logit contributions | |
LLOW | +2.105 |
tten | +2.043 |
�� | +1.995 |
Swordsman | +1.940 |
FACE | +1.913 |
Neg logit contributions | |
ippers | -2.177 |
ipper | -2.116 |
rones | -1.721 |
lems | -1.721 |
mast | -1.694 |
Token | this |
Token ID | 428 |
Feature activation | -8.16e-01 |
Pos logit contributions | |
LLOW | +0.374 |
tten | +0.366 |
�� | +0.360 |
Swordsman | +0.351 |
FACE | +0.347 |
Neg logit contributions | |
ippers | -0.265 |
ipper | -0.258 |
lems | -0.195 |
rones | -0.195 |
lihood | -0.192 |
Token | weekend |
Token ID | 5041 |
Feature activation | +1.77e+00 |
Pos logit contributions | |
ippers | +0.746 |
ipper | +0.732 |
rones | +0.614 |
lems | +0.612 |
Carney | +0.605 |
Neg logit contributions | |
LLOW | -0.487 |
tten | -0.471 |
�� | -0.459 |
Swordsman | -0.440 |
FACE | -0.432 |
Token | in |
Token ID | 287 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
LLOW | +1.453 |
tten | +1.412 |
�� | +1.389 |
Swordsman | +1.358 |
FACE | +1.338 |
Neg logit contributions | |
ippers | -1.209 |
ipper | -1.176 |
lems | -0.920 |
rones | -0.920 |
mast | -0.902 |
Token | Wisconsin |
Token ID | 9279 |
Feature activation | +7.10e+00 |
Pos logit contributions | |
LLOW | +0.329 |
tten | +0.320 |
�� | +0.313 |
Swordsman | +0.305 |
FACE | +0.301 |
Neg logit contributions | |
ippers | -0.262 |
ipper | -0.255 |
rones | -0.198 |
lems | -0.197 |
mast | -0.195 |
Token | , |
Token ID | 11 |
Feature activation | +7.72e-01 |
Pos logit contributions | |
LLOW | +5.044 |
tten | +4.846 |
�� | +4.765 |
Swordsman | +4.607 |
FACE | +4.563 |
Neg logit contributions | |
ippers | -5.430 |
ipper | -5.318 |
lems | -4.329 |
rones | -4.322 |
mast | -4.278 |
Token | another |
Token ID | 1194 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
LLOW | +0.590 |
tten | +0.573 |
�� | +0.560 |
Swordsman | +0.544 |
FACE | +0.537 |
Neg logit contributions | |
ippers | -0.575 |
ipper | -0.562 |
rones | -0.453 |
lems | -0.452 |
mast | -0.444 |
Token | state |
Token ID | 1181 |
Feature activation | +8.44e-01 |
Pos logit contributions | |
ippers | +0.789 |
ipper | +0.771 |
rones | +0.625 |
lems | +0.624 |
Carney | +0.613 |
Neg logit contributions | |
LLOW | -0.742 |
tten | -0.723 |
�� | -0.707 |
Swordsman | -0.684 |
FACE | -0.674 |
Token | with |
Token ID | 351 |
Feature activation | +3.42e-01 |
Pos logit contributions | |
LLOW | +0.635 |
tten | +0.617 |
�� | +0.605 |
Swordsman | +0.585 |
FACE | +0.577 |
Neg logit contributions | |
ippers | -0.639 |
ipper | -0.625 |
rones | -0.505 |
lems | -0.503 |
mast | -0.495 |
Token | a |
Token ID | 257 |
Feature activation | +1.37e-01 |
Pos logit contributions | |
LLOW | +0.267 |
tten | +0.261 |
�� | +0.255 |
Swordsman | +0.247 |
FACE | +0.244 |
Neg logit contributions | |
ippers | -0.248 |
ipper | -0.244 |
rones | -0.194 |
lems | -0.193 |
mast | -0.191 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
ippers | +1.563 |
ipper | +1.529 |
rones | +1.225 |
lems | +1.221 |
Carney | +1.210 |
Neg logit contributions | |
LLOW | -1.643 |
tten | -1.611 |
�� | -1.576 |
Swordsman | -1.526 |
ADRA | -1.525 |
Token | paid |
Token ID | 3432 |
Feature activation | +2.01e+00 |
Pos logit contributions | |
LLOW | +1.788 |
tten | +1.721 |
�� | +1.672 |
Swordsman | +1.644 |
FACE | +1.622 |
Neg logit contributions | |
ippers | -1.774 |
ipper | -1.738 |
rones | -1.402 |
lihood | -1.399 |
lems | -1.385 |
Token | for |
Token ID | 329 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
LLOW | +1.710 |
tten | +1.663 |
�� | +1.607 |
Swordsman | +1.580 |
FACE | +1.574 |
Neg logit contributions | |
ippers | -1.305 |
ipper | -1.271 |
lihood | -0.998 |
lems | -0.983 |
rones | -0.983 |
Token | by |
Token ID | 416 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
LLOW | +1.767 |
tten | +1.709 |
�� | +1.670 |
Swordsman | +1.665 |
FACE | +1.631 |
Neg logit contributions | |
ippers | -1.235 |
ipper | -1.190 |
lems | -0.910 |
ploy | -0.907 |
rones | -0.894 |
Token | George |
Token ID | 4502 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
ippers | +0.137 |
ipper | +0.134 |
lems | +0.108 |
rones | +0.108 |
lihood | +0.106 |
Neg logit contributions | |
LLOW | -0.131 |
tten | -0.128 |
�� | -0.124 |
Swordsman | -0.121 |
FACE | -0.119 |
Token | Soros |
Token ID | 27900 |
Feature activation | +7.07e+00 |
Pos logit contributions | |
ippers | +0.149 |
ipper | +0.147 |
rones | +0.121 |
lems | +0.121 |
lihood | +0.119 |
Neg logit contributions | |
LLOW | -0.119 |
tten | -0.115 |
�� | -0.112 |
Swordsman | -0.109 |
FACE | -0.107 |
Token | 's |
Token ID | 338 |
Feature activation | +5.53e-01 |
Pos logit contributions | |
LLOW | +4.756 |
tten | +4.487 |
Swordsman | +4.325 |
�� | +4.270 |
FACE | +4.229 |
Neg logit contributions | |
ippers | -5.647 |
ipper | -5.506 |
lems | -4.557 |
rones | -4.510 |
lihood | -4.500 |
Token | pre |
Token ID | 662 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
LLOW | +0.448 |
tten | +0.435 |
�� | +0.424 |
Swordsman | +0.415 |
FACE | +0.408 |
Neg logit contributions | |
ippers | -0.388 |
ipper | -0.379 |
rones | -0.299 |
lems | -0.298 |
mast | -0.293 |
Token | - |
Token ID | 12 |
Feature activation | +1.39e+00 |
Pos logit contributions | |
LLOW | +2.187 |
tten | +2.173 |
�� | +2.088 |
FACE | +2.038 |
Swordsman | +2.023 |
Neg logit contributions | |
ippers | -1.985 |
ipper | -1.921 |
lems | -1.539 |
lihood | -1.519 |
rones | -1.513 |
Token | tax |
Token ID | 19290 |
Feature activation | -8.35e-01 |
Pos logit contributions | |
LLOW | +1.021 |
tten | +1.004 |
�� | +0.970 |
FACE | +0.940 |
Swordsman | +0.932 |
Neg logit contributions | |
ippers | -1.079 |
ipper | -1.051 |
lems | -0.851 |
lihood | -0.846 |
rones | -0.844 |
Token | dollars |
Token ID | 5054 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ippers | +0.597 |
ipper | +0.583 |
rones | +0.463 |
lems | +0.461 |
Carney | +0.457 |
Neg logit contributions | |
LLOW | -0.664 |
tten | -0.648 |
�� | -0.635 |
Swordsman | -0.615 |
ADRA | -0.609 |
Token | . |
Token ID | 13 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
LLOW | +1.390 |
tten | +1.369 |
�� | +1.328 |
Swordsman | +1.279 |
ADRA | +1.260 |
Neg logit contributions | |
ippers | -1.421 |
ipper | -1.388 |
Carney | -1.119 |
rones | -1.115 |
lems | -1.110 |
Token | Fro |
Token ID | 9734 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
LLOW | +1.420 |
tten | +1.358 |
Swordsman | +1.323 |
�� | +1.310 |
FACE | +1.287 |
Neg logit contributions | |
ippers | -1.491 |
ipper | -1.455 |
rones | -1.190 |
ploy | -1.168 |
lems | -1.165 |
Token | do |
Token ID | 4598 |
Feature activation | +2.85e+00 |
Pos logit contributions | |
LLOW | +1.561 |
tten | +1.543 |
�� | +1.459 |
Swordsman | +1.429 |
FACE | +1.428 |
Neg logit contributions | |
ippers | -1.252 |
ipper | -1.206 |
rones | -0.960 |
Carney | -0.958 |
lihood | -0.945 |
Token | falls |
Token ID | 8953 |
Feature activation | +3.50e+00 |
Pos logit contributions | |
LLOW | +2.183 |
tten | +2.147 |
�� | +2.076 |
Swordsman | +2.058 |
FACE | +2.005 |
Neg logit contributions | |
ippers | -2.075 |
ipper | -2.026 |
rones | -1.612 |
lems | -1.598 |
ploy | -1.585 |
Token | to |
Token ID | 284 |
Feature activation | +4.31e+00 |
Pos logit contributions | |
LLOW | +2.523 |
tten | +2.474 |
�� | +2.354 |
Swordsman | +2.350 |
FACE | +2.301 |
Neg logit contributions | |
ippers | -2.700 |
ipper | -2.623 |
rones | -2.156 |
Carney | -2.129 |
ploy | -2.102 |
Token | Shel |
Token ID | 15325 |
Feature activation | +7.06e+00 |
Pos logit contributions | |
LLOW | +3.068 |
tten | +3.050 |
�� | +2.884 |
Swordsman | +2.871 |
ADRA | +2.798 |
Neg logit contributions | |
ippers | -3.373 |
ipper | -3.303 |
rones | -2.717 |
Carney | -2.700 |
ploy | -2.664 |
Token | ob |
Token ID | 672 |
Feature activation | +1.17e+00 |
Pos logit contributions | |
tten | +6.559 |
LLOW | +6.534 |
Swordsman | +6.178 |
icide | +5.979 |
ATH | +5.965 |
Neg logit contributions | |
ippers | -3.481 |
ipper | -3.405 |
Carney | -2.711 |
lihood | -2.655 |
rones | -2.517 |
Token | 's |
Token ID | 338 |
Feature activation | +4.98e-01 |
Pos logit contributions | |
LLOW | +0.920 |
tten | +0.899 |
�� | +0.880 |
Swordsman | +0.852 |
ADRA | +0.841 |
Neg logit contributions | |
ippers | -0.851 |
ipper | -0.831 |
rones | -0.661 |
lems | -0.658 |
Carney | -0.652 |
Token | sting |
Token ID | 24276 |
Feature activation | +7.86e-01 |
Pos logit contributions | |
LLOW | +0.372 |
tten | +0.363 |
�� | +0.355 |
Swordsman | +0.345 |
ADRA | +0.339 |
Neg logit contributions | |
ippers | -0.379 |
ipper | -0.370 |
rones | -0.300 |
Carney | -0.297 |
lems | -0.297 |
Token | . |
Token ID | 13 |
Feature activation | +2.67e-01 |
Pos logit contributions | |
LLOW | +0.626 |
tten | +0.611 |
�� | +0.598 |
Swordsman | +0.580 |
ADRA | +0.572 |
Neg logit contributions | |
ippers | -0.561 |
ipper | -0.547 |
rones | -0.434 |
lems | -0.432 |
Carney | -0.429 |
Token | But |
Token ID | 887 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
LLOW | +0.202 |
tten | +0.196 |
�� | +0.190 |
Swordsman | +0.188 |
FACE | +0.184 |
Neg logit contributions | |
ippers | -0.202 |
ipper | -0.197 |
rones | -0.159 |
lems | -0.157 |
ploy | -0.156 |
Token | Before |
Token ID | 8421 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
ippers | +0.350 |
ipper | +0.343 |
rones | +0.285 |
lems | +0.284 |
Carney | +0.282 |
Neg logit contributions | |
LLOW | -0.260 |
tten | -0.252 |
�� | -0.245 |
Swordsman | -0.236 |
ADRA | -0.233 |
Token | the |
Token ID | 262 |
Feature activation | +6.18e-01 |
Pos logit contributions | |
LLOW | +2.722 |
tten | +2.680 |
�� | +2.551 |
Swordsman | +2.530 |
FACE | +2.483 |
Neg logit contributions | |
ippers | -3.043 |
ipper | -2.965 |
rones | -2.472 |
lems | -2.454 |
mast | -2.403 |
Token | gameplay |
Token ID | 11327 |
Feature activation | +3.20e+00 |
Pos logit contributions | |
LLOW | +0.447 |
tten | +0.436 |
�� | +0.424 |
Swordsman | +0.414 |
FACE | +0.405 |
Neg logit contributions | |
ippers | -0.484 |
ipper | -0.474 |
rones | -0.387 |
lems | -0.385 |
mast | -0.379 |
Token | proved |
Token ID | 8302 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
LLOW | +2.474 |
tten | +2.451 |
�� | +2.313 |
Swordsman | +2.303 |
FACE | +2.264 |
Neg logit contributions | |
ippers | -2.327 |
ipper | -2.272 |
rones | -1.836 |
lems | -1.820 |
ploy | -1.799 |
Token | Donkey |
Token ID | 43823 |
Feature activation | +2.90e+00 |
Pos logit contributions | |
ippers | +0.904 |
ipper | +0.884 |
rones | +0.711 |
lems | +0.710 |
Carney | +0.704 |
Neg logit contributions | |
LLOW | -0.896 |
tten | -0.873 |
�� | -0.855 |
Swordsman | -0.825 |
ADRA | -0.816 |
Token | Kong |
Token ID | 9071 |
Feature activation | +7.03e+00 |
Pos logit contributions | |
LLOW | +0.815 |
tten | +0.747 |
�� | +0.686 |
Swordsman | +0.656 |
FACE | +0.628 |
Neg logit contributions | |
ippers | -3.543 |
ipper | -3.491 |
lems | -3.084 |
rones | -3.081 |
lihood | -3.073 |
Token | Country |
Token ID | 12946 |
Feature activation | +3.51e+00 |
Pos logit contributions | |
LLOW | +4.089 |
tten | +3.976 |
Swordsman | +3.808 |
�� | +3.715 |
ADRA | +3.658 |
Neg logit contributions | |
ippers | -6.189 |
ipper | -6.068 |
rones | -5.194 |
lems | -5.146 |
mast | -5.107 |
Token | 2 |
Token ID | 362 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
LLOW | +2.549 |
tten | +2.488 |
�� | +2.409 |
Swordsman | +2.350 |
ADRA | +2.300 |
Neg logit contributions | |
ippers | -2.720 |
ipper | -2.664 |
rones | -2.163 |
lems | -2.146 |
mast | -2.135 |
Token | : |
Token ID | 25 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
LLOW | +1.273 |
tten | +1.247 |
�� | +1.212 |
Swordsman | +1.178 |
FACE | +1.160 |
Neg logit contributions | |
ippers | -1.211 |
ipper | -1.184 |
rones | -0.947 |
lems | -0.944 |
Carney | -0.933 |
Token | D |
Token ID | 360 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
LLOW | +1.555 |
tten | +1.505 |
�� | +1.468 |
Swordsman | +1.419 |
ADRA | +1.394 |
Neg logit contributions | |
ippers | -2.006 |
ipper | -1.967 |
rones | -1.627 |
lems | -1.620 |
Carney | -1.609 |
Token | iddy |
Token ID | 34208 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
LLOW | +0.876 |
tten | +0.854 |
�� | +0.834 |
Swordsman | +0.805 |
FACE | +0.796 |
Neg logit contributions | |
ippers | -0.905 |
ipper | -0.884 |
rones | -0.714 |
lems | -0.712 |
Carney | -0.707 |
Token | release |
Token ID | 2650 |
Feature activation | +1.51e+00 |
Pos logit contributions | |
LLOW | +0.080 |
tten | +0.078 |
�� | +0.077 |
Swordsman | +0.074 |
FACE | +0.073 |
Neg logit contributions | |
ippers | -0.073 |
ipper | -0.072 |
rones | -0.057 |
lems | -0.057 |
Carney | -0.056 |
Token | date |
Token ID | 3128 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
LLOW | +1.307 |
tten | +1.282 |
�� | +1.241 |
Swordsman | +1.226 |
FACE | +1.195 |
Neg logit contributions | |
ippers | -0.963 |
ipper | -0.938 |
rones | -0.733 |
lems | -0.729 |
Carney | -0.711 |
Token | for |
Token ID | 329 |
Feature activation | +2.96e+00 |
Pos logit contributions | |
LLOW | +1.746 |
tten | +1.703 |
�� | +1.640 |
Swordsman | +1.629 |
FACE | +1.586 |
Neg logit contributions | |
ippers | -1.544 |
ipper | -1.511 |
rones | -1.209 |
lems | -1.200 |
Carney | -1.171 |
Token | " |
Token ID | 366 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
LLOW | +2.095 |
tten | +1.999 |
�� | +1.950 |
Swordsman | +1.926 |
FACE | +1.865 |
Neg logit contributions | |
ippers | -2.370 |
ipper | -2.310 |
rones | -1.921 |
lems | -1.899 |
ploy | -1.874 |
Token | Final |
Token ID | 19006 |
Feature activation | +2.27e+00 |
Pos logit contributions | |
LLOW | +0.989 |
tten | +0.948 |
�� | +0.923 |
Swordsman | +0.881 |
FACE | +0.874 |
Neg logit contributions | |
ippers | -1.711 |
ipper | -1.679 |
rones | -1.425 |
lems | -1.417 |
Carney | -1.413 |
Token | Fantasy |
Token ID | 10982 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
LLOW | +1.822 |
tten | +1.778 |
�� | +1.743 |
Swordsman | +1.733 |
FACE | +1.674 |
Neg logit contributions | |
ippers | -1.598 |
ipper | -1.557 |
rones | -1.246 |
lems | -1.224 |
Carney | -1.202 |
Token | V |
Token ID | 569 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
LLOW | +4.877 |
tten | +4.780 |
Swordsman | +4.623 |
�� | +4.546 |
Bleach | +4.399 |
Neg logit contributions | |
ippers | -5.521 |
ipper | -5.369 |
rones | -4.487 |
lems | -4.453 |
ploy | -4.267 |
Token | X |
Token ID | 55 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
LLOW | +2.636 |
tten | +2.554 |
�� | +2.458 |
Swordsman | +2.394 |
FACE | +2.350 |
Neg logit contributions | |
ippers | -3.521 |
ipper | -3.456 |
lems | -2.906 |
rones | -2.906 |
Carney | -2.875 |
Token | " |
Token ID | 1 |
Feature activation | +6.30e-01 |
Pos logit contributions | |
LLOW | +0.740 |
tten | +0.717 |
�� | +0.700 |
Swordsman | +0.673 |
FACE | +0.662 |
Neg logit contributions | |
ippers | -1.019 |
ipper | -0.998 |
rones | -0.832 |
lems | -0.831 |
Carney | -0.821 |
Token | will |
Token ID | 481 |
Feature activation | +1.70e-01 |
Pos logit contributions | |
LLOW | +0.509 |
tten | +0.497 |
�� | +0.485 |
Swordsman | +0.473 |
FACE | +0.465 |
Neg logit contributions | |
ippers | -0.441 |
ipper | -0.430 |
rones | -0.340 |
lems | -0.339 |
Carney | -0.335 |
Token | be |
Token ID | 307 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
LLOW | +0.131 |
tten | +0.128 |
�� | +0.124 |
Swordsman | +0.121 |
FACE | +0.119 |
Neg logit contributions | |
ippers | -0.125 |
ipper | -0.122 |
rones | -0.098 |
lems | -0.098 |
Carney | -0.097 |
Token | 9 |
Token ID | 24 |
Feature activation | +6.77e-01 |
Pos logit contributions | |
ippers | +1.557 |
ipper | +1.523 |
rones | +1.219 |
lems | +1.214 |
Carney | +1.204 |
Neg logit contributions | |
LLOW | -1.622 |
tten | -1.584 |
�� | -1.555 |
Swordsman | -1.498 |
ADRA | -1.483 |
Token | million |
Token ID | 1510 |
Feature activation | +2.39e+00 |
Pos logit contributions | |
LLOW | +0.430 |
tten | +0.417 |
�� | +0.406 |
Swordsman | +0.390 |
ADRA | +0.384 |
Neg logit contributions | |
ippers | -0.594 |
ipper | -0.582 |
rones | -0.484 |
lems | -0.483 |
Carney | -0.479 |
Token | in |
Token ID | 287 |
Feature activation | +1.54e+00 |
Pos logit contributions | |
LLOW | +1.821 |
tten | +1.761 |
�� | +1.723 |
Swordsman | +1.695 |
FACE | +1.675 |
Neg logit contributions | |
ippers | -1.789 |
ipper | -1.744 |
rones | -1.390 |
lems | -1.387 |
lihood | -1.377 |
Token | April |
Token ID | 3035 |
Feature activation | -2.92e+00 |
Pos logit contributions | |
LLOW | +1.147 |
tten | +1.112 |
�� | +1.089 |
Swordsman | +1.060 |
FACE | +1.053 |
Neg logit contributions | |
ippers | -1.176 |
ipper | -1.150 |
lems | -0.924 |
rones | -0.922 |
lihood | -0.919 |
Token | , |
Token ID | 11 |
Feature activation | -2.33e-01 |
Pos logit contributions | |
ippers | +1.954 |
ipper | +1.903 |
rones | +1.510 |
Carney | +1.499 |
lems | +1.493 |
Neg logit contributions | |
LLOW | -2.418 |
tten | -2.389 |
�� | -2.315 |
ADRA | -2.265 |
Swordsman | -2.246 |
Token | narrowly |
Token ID | 26167 |
Feature activation | +6.97e+00 |
Pos logit contributions | |
ippers | +0.184 |
ipper | +0.180 |
rones | +0.147 |
lems | +0.146 |
Carney | +0.145 |
Neg logit contributions | |
LLOW | -0.168 |
tten | -0.164 |
�� | -0.160 |
Swordsman | -0.154 |
ADRA | -0.152 |
Token | in |
Token ID | 287 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
tten | +3.977 |
LLOW | +3.976 |
�� | +3.704 |
Swordsman | +3.691 |
Bleach | +3.631 |
Neg logit contributions | |
ippers | -6.137 |
ipper | -6.082 |
rones | -5.183 |
lems | -5.109 |
ploy | -5.008 |
Token | ching |
Token ID | 10813 |
Feature activation | +1.20e+00 |
Pos logit contributions | |
LLOW | +1.424 |
tten | +1.392 |
�� | +1.354 |
Swordsman | +1.303 |
FACE | +1.294 |
Neg logit contributions | |
ippers | -1.292 |
ipper | -1.263 |
rones | -1.004 |
lems | -0.998 |
lihood | -0.991 |
Token | out |
Token ID | 503 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
LLOW | +0.862 |
tten | +0.840 |
�� | +0.810 |
Swordsman | +0.800 |
FACE | +0.793 |
Neg logit contributions | |
ippers | -0.944 |
ipper | -0.925 |
rones | -0.751 |
lems | -0.745 |
ploy | -0.740 |
Token | Mrs |
Token ID | 9074 |
Feature activation | -4.16e+00 |
Pos logit contributions | |
LLOW | +0.923 |
tten | +0.904 |
�� | +0.877 |
Swordsman | +0.859 |
FACE | +0.844 |
Neg logit contributions | |
ippers | -0.973 |
ipper | -0.952 |
rones | -0.773 |
lems | -0.767 |
ploy | -0.755 |
Token | . |
Token ID | 13 |
Feature activation | -1.70e+00 |
Pos logit contributions | |
ippers | +2.897 |
ipper | +2.826 |
Carney | +2.244 |
lems | +2.230 |
rones | +2.220 |
Neg logit contributions | |
LLOW | -3.352 |
tten | -3.290 |
�� | -3.286 |
ADRA | -3.185 |
Swordsman | -3.095 |
Token | risk |
Token ID | 2526 |
Feature activation | +6.08e-01 |
Pos logit contributions | |
LLOW | +0.612 |
tten | +0.597 |
�� | +0.584 |
FACE | +0.564 |
Swordsman | +0.561 |
Neg logit contributions | |
ippers | -0.513 |
ipper | -0.498 |
lems | -0.394 |
rones | -0.394 |
Carney | -0.385 |
Token | of |
Token ID | 286 |
Feature activation | +6.49e-01 |
Pos logit contributions | |
LLOW | +0.480 |
tten | +0.468 |
�� | +0.458 |
Swordsman | +0.443 |
FACE | +0.439 |
Neg logit contributions | |
ippers | -0.438 |
ipper | -0.428 |
rones | -0.341 |
lems | -0.340 |
Carney | -0.335 |
Token | the |
Token ID | 262 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
LLOW | +0.436 |
tten | +0.425 |
�� | +0.414 |
Swordsman | +0.398 |
ADRA | +0.392 |
Neg logit contributions | |
ippers | -0.543 |
ipper | -0.532 |
rones | -0.439 |
lems | -0.439 |
Carney | -0.432 |
Token | situation |
Token ID | 3074 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
LLOW | +0.690 |
tten | +0.669 |
�� | +0.652 |
Swordsman | +0.626 |
ADRA | +0.617 |
Neg logit contributions | |
ippers | -0.923 |
ipper | -0.904 |
rones | -0.753 |
lems | -0.749 |
Carney | -0.740 |
Token | in |
Token ID | 287 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
LLOW | +2.054 |
tten | +2.004 |
�� | +1.958 |
Swordsman | +1.882 |
FACE | +1.870 |
Neg logit contributions | |
ippers | -2.141 |
ipper | -2.082 |
rones | -1.704 |
lems | -1.700 |
lihood | -1.667 |
Token | Chennai |
Token ID | 47678 |
Feature activation | +6.88e+00 |
Pos logit contributions | |
LLOW | +1.312 |
tten | +1.272 |
�� | +1.231 |
Swordsman | +1.188 |
ADRA | +1.171 |
Neg logit contributions | |
ippers | -1.557 |
ipper | -1.525 |
rones | -1.264 |
lems | -1.259 |
lihood | -1.239 |
Token | becoming |
Token ID | 5033 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
LLOW | +4.914 |
tten | +4.738 |
�� | +4.646 |
Swordsman | +4.474 |
FACE | +4.433 |
Neg logit contributions | |
ippers | -5.177 |
ipper | -5.041 |
lihood | -4.185 |
rones | -4.175 |
lems | -4.170 |
Token | even |
Token ID | 772 |
Feature activation | -8.30e-01 |
Pos logit contributions | |
LLOW | +1.363 |
tten | +1.344 |
�� | +1.314 |
Swordsman | +1.250 |
FACE | +1.248 |
Neg logit contributions | |
ippers | -1.483 |
ipper | -1.437 |
rones | -1.188 |
lems | -1.171 |
lihood | -1.158 |
Token | worse |
Token ID | 4785 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
ippers | +0.677 |
ipper | +0.661 |
rones | +0.543 |
lems | +0.541 |
Carney | +0.535 |
Neg logit contributions | |
LLOW | -0.577 |
tten | -0.563 |
�� | -0.549 |
Swordsman | -0.528 |
FACE | -0.523 |
Token | over |
Token ID | 625 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
LLOW | +0.452 |
tten | +0.439 |
�� | +0.428 |
Swordsman | +0.414 |
FACE | +0.410 |
Neg logit contributions | |
ippers | -0.480 |
ipper | -0.469 |
rones | -0.380 |
lems | -0.379 |
Carney | -0.374 |
Token | the |
Token ID | 262 |
Feature activation | +3.89e-01 |
Pos logit contributions | |
LLOW | +1.097 |
tten | +1.069 |
�� | +1.049 |
Swordsman | +1.015 |
ADRA | +1.004 |
Neg logit contributions | |
ippers | -1.038 |
ipper | -1.014 |
rones | -0.809 |
lems | -0.807 |
Carney | -0.801 |
Token | to |
Token ID | 284 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
LLOW | +0.526 |
tten | +0.510 |
�� | +0.499 |
Swordsman | +0.484 |
FACE | +0.479 |
Neg logit contributions | |
ippers | -0.539 |
ipper | -0.526 |
lems | -0.425 |
rones | -0.424 |
Carney | -0.419 |
Token | Recap |
Token ID | 46585 |
Feature activation | +1.30e+00 |
Pos logit contributions | |
LLOW | +0.864 |
tten | +0.838 |
�� | +0.806 |
FACE | +0.795 |
Swordsman | +0.792 |
Neg logit contributions | |
ippers | -0.785 |
ipper | -0.767 |
lems | -0.616 |
rones | -0.608 |
lihood | -0.605 |
Token | San |
Token ID | 2986 |
Feature activation | -1.07e+00 |
Pos logit contributions | |
LLOW | +1.024 |
tten | +0.989 |
�� | +0.969 |
Swordsman | +0.941 |
FACE | +0.934 |
Neg logit contributions | |
ippers | -0.938 |
ipper | -0.913 |
lems | -0.731 |
rones | -0.727 |
lihood | -0.718 |
Token | Juan |
Token ID | 16852 |
Feature activation | +3.90e-01 |
Pos logit contributions | |
ippers | +0.877 |
ipper | +0.858 |
lems | +0.703 |
rones | +0.700 |
Carney | +0.695 |
Neg logit contributions | |
LLOW | -0.738 |
tten | -0.718 |
�� | -0.702 |
Swordsman | -0.675 |
FACE | -0.666 |
Token | Del |
Token ID | 4216 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
LLOW | +0.332 |
tten | +0.323 |
�� | +0.317 |
Swordsman | +0.310 |
ADRA | +0.306 |
Neg logit contributions | |
ippers | -0.257 |
ipper | -0.250 |
lems | -0.195 |
rones | -0.193 |
lihood | -0.190 |
Token | Sur |
Token ID | 4198 |
Feature activation | +6.87e+00 |
Pos logit contributions | |
LLOW | +0.855 |
tten | +0.832 |
�� | +0.801 |
Swordsman | +0.793 |
FACE | +0.781 |
Neg logit contributions | |
ippers | -0.866 |
ipper | -0.844 |
lems | -0.684 |
rones | -0.677 |
lihood | -0.670 |
Token | Episode |
Token ID | 7922 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
LLOW | +5.072 |
tten | +4.777 |
�� | +4.684 |
FACE | +4.636 |
Swordsman | +4.570 |
Neg logit contributions | |
ippers | -5.127 |
ipper | -4.938 |
lems | -4.018 |
ploy | -3.993 |
lihood | -3.950 |
Token | 5 |
Token ID | 642 |
Feature activation | +9.03e-01 |
Pos logit contributions | |
ippers | +1.168 |
ipper | +1.142 |
rones | +0.920 |
lems | +0.915 |
Carney | +0.910 |
Neg logit contributions | |
LLOW | -1.151 |
tten | -1.123 |
�� | -1.100 |
Swordsman | -1.062 |
ADRA | -1.052 |
Token |
|
Token ID | 198 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
LLOW | +0.694 |
tten | +0.670 |
�� | +0.655 |
Swordsman | +0.638 |
FACE | +0.631 |
Neg logit contributions | |
ippers | -0.672 |
ipper | -0.655 |
lems | -0.528 |
rones | -0.525 |
lihood | -0.522 |
Token |
|
Token ID | 198 |
Feature activation | -8.32e-02 |
Pos logit contributions | |
ippers | +0.185 |
ipper | +0.180 |
rones | +0.142 |
lems | +0.141 |
Carney | +0.139 |
Neg logit contributions | |
LLOW | -0.220 |
tten | -0.214 |
�� | -0.209 |
Swordsman | -0.203 |
FACE | -0.202 |
Token | Coach |
Token ID | 40677 |
Feature activation | +2.13e+00 |
Pos logit contributions | |
ippers | +0.070 |
ipper | +0.069 |
lems | +0.057 |
rones | +0.057 |
Carney | +0.057 |
Neg logit contributions | |
LLOW | -0.056 |
tten | -0.053 |
�� | -0.052 |
FACE | -0.051 |
Swordsman | -0.050 |
Token | details |
Token ID | 3307 |
Feature activation | +2.72e+00 |
Pos logit contributions | |
LLOW | +0.413 |
tten | +0.400 |
�� | +0.387 |
Swordsman | +0.377 |
FACE | +0.371 |
Neg logit contributions | |
ippers | -0.475 |
ipper | -0.464 |
rones | -0.380 |
lems | -0.378 |
ploy | -0.374 |
Token | and |
Token ID | 290 |
Feature activation | +4.83e-01 |
Pos logit contributions | |
LLOW | +1.914 |
tten | +1.849 |
�� | +1.804 |
Swordsman | +1.744 |
FACE | +1.719 |
Neg logit contributions | |
ippers | -2.185 |
ipper | -2.136 |
rones | -1.749 |
lems | -1.746 |
lihood | -1.732 |
Token | analysis |
Token ID | 3781 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
LLOW | +0.373 |
tten | +0.362 |
�� | +0.353 |
Swordsman | +0.344 |
FACE | +0.340 |
Neg logit contributions | |
ippers | -0.356 |
ipper | -0.349 |
rones | -0.279 |
lems | -0.278 |
ploy | -0.273 |
Token | . |
Token ID | 13 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
LLOW | +2.004 |
tten | +1.938 |
�� | +1.886 |
Swordsman | +1.820 |
FACE | +1.799 |
Neg logit contributions | |
ippers | -2.266 |
ipper | -2.218 |
rones | -1.810 |
lems | -1.807 |
lihood | -1.792 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +4.01e+00 |
Pos logit contributions | |
LLOW | +1.152 |
tten | +1.104 |
�� | +1.080 |
Swordsman | +1.053 |
FACE | +1.039 |
Neg logit contributions | |
ippers | -1.218 |
ipper | -1.193 |
rones | -0.967 |
lems | -0.962 |
lihood | -0.954 |
Token | Sym |
Token ID | 43094 |
Feature activation | +6.87e+00 |
Pos logit contributions | |
LLOW | +2.676 |
tten | +2.548 |
FACE | +2.454 |
�� | +2.440 |
ATH | +2.413 |
Neg logit contributions | |
ippers | -3.338 |
ipper | -3.272 |
Carney | -2.730 |
rones | -2.706 |
lihood | -2.702 |
Token | one |
Token ID | 505 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
tten | +4.829 |
LLOW | +4.826 |
�� | +4.554 |
Swordsman | +4.510 |
FACE | +4.496 |
Neg logit contributions | |
ippers | -5.300 |
ipper | -5.155 |
lems | -4.165 |
Carney | -4.164 |
rones | -4.141 |
Token | Sanders |
Token ID | 5831 |
Feature activation | +2.39e-01 |
Pos logit contributions | |
ippers | +2.077 |
ipper | +2.031 |
rones | +1.637 |
lems | +1.631 |
Carney | +1.618 |
Neg logit contributions | |
LLOW | -2.047 |
tten | -1.994 |
�� | -1.953 |
Swordsman | -1.886 |
ADRA | -1.865 |
Token | was |
Token ID | 373 |
Feature activation | -5.12e-03 |
Pos logit contributions | |
LLOW | +0.187 |
tten | +0.183 |
�� | +0.179 |
Swordsman | +0.174 |
FACE | +0.172 |
Neg logit contributions | |
ippers | -0.173 |
ipper | -0.169 |
rones | -0.135 |
lems | -0.134 |
lihood | -0.132 |
Token | not |
Token ID | 407 |
Feature activation | +4.33e-01 |
Pos logit contributions | |
ippers | +0.004 |
ipper | +0.004 |
rones | +0.003 |
lems | +0.003 |
lihood | +0.003 |
Neg logit contributions | |
LLOW | -0.004 |
tten | -0.004 |
�� | -0.004 |
Swordsman | -0.004 |
FACE | -0.004 |
Token | having |
Token ID | 1719 |
Feature activation | -1.47e-01 |
Pos logit contributions | |
LLOW | +0.336 |
tten | +0.329 |
�� | +0.320 |
Swordsman | +0.311 |
FACE | +0.308 |
Neg logit contributions | |
ippers | -0.318 |
ipper | -0.310 |
rones | -0.249 |
lems | -0.247 |
lihood | -0.242 |
Token | at |
Token ID | 379 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
LLOW | +0.974 |
tten | +0.951 |
�� | +0.919 |
Swordsman | +0.901 |
FACE | +0.888 |
Neg logit contributions | |
ippers | -0.907 |
ipper | -0.890 |
rones | -0.711 |
lems | -0.705 |
mast | -0.693 |
Token | town |
Token ID | 3240 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
LLOW | +0.963 |
tten | +0.939 |
�� | +0.916 |
Swordsman | +0.892 |
FACE | +0.877 |
Neg logit contributions | |
ippers | -0.954 |
ipper | -0.934 |
rones | -0.749 |
lems | -0.747 |
lihood | -0.736 |
Token | meetings |
Token ID | 8292 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
LLOW | +0.964 |
tten | +0.940 |
�� | +0.928 |
Swordsman | +0.893 |
FACE | +0.890 |
Neg logit contributions | |
ippers | -0.615 |
ipper | -0.603 |
lems | -0.449 |
rones | -0.448 |
lihood | -0.442 |
Token | in |
Token ID | 287 |
Feature activation | +6.67e-01 |
Pos logit contributions | |
LLOW | +0.892 |
tten | +0.873 |
�� | +0.843 |
Swordsman | +0.825 |
FACE | +0.810 |
Neg logit contributions | |
ippers | -0.830 |
ipper | -0.814 |
lems | -0.645 |
rones | -0.642 |
mast | -0.631 |
Token | New |
Token ID | 968 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
LLOW | +0.530 |
tten | +0.517 |
�� | +0.503 |
Swordsman | +0.490 |
FACE | +0.482 |
Neg logit contributions | |
ippers | -0.480 |
ipper | -0.469 |
lems | -0.372 |
rones | -0.372 |
mast | -0.366 |
Token | Hampshire |
Token ID | 13910 |
Feature activation | +6.85e+00 |
Pos logit contributions | |
ippers | +0.670 |
ipper | +0.652 |
rones | +0.503 |
lems | +0.502 |
Carney | +0.501 |
Neg logit contributions | |
LLOW | -0.911 |
tten | -0.886 |
�� | -0.873 |
Swordsman | -0.846 |
ADRA | -0.843 |
Token | and |
Token ID | 290 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
LLOW | +4.972 |
tten | +4.868 |
Swordsman | +4.618 |
�� | +4.567 |
FACE | +4.510 |
Neg logit contributions | |
ippers | -5.010 |
ipper | -4.949 |
lems | -3.976 |
rones | -3.928 |
lihood | -3.897 |
Token | Iowa |
Token ID | 9406 |
Feature activation | +4.35e+00 |
Pos logit contributions | |
LLOW | +0.884 |
tten | +0.864 |
�� | +0.840 |
Swordsman | +0.820 |
FACE | +0.807 |
Neg logit contributions | |
ippers | -0.815 |
ipper | -0.798 |
rones | -0.634 |
lems | -0.633 |
mast | -0.620 |
Token | are |
Token ID | 389 |
Feature activation | +7.53e-01 |
Pos logit contributions | |
LLOW | +3.331 |
tten | +3.272 |
�� | +3.154 |
Swordsman | +3.069 |
FACE | +3.043 |
Neg logit contributions | |
ippers | -3.119 |
ipper | -3.081 |
lems | -2.433 |
rones | -2.405 |
lihood | -2.392 |
Token | sending |
Token ID | 7216 |
Feature activation | +9.84e-01 |
Pos logit contributions | |
LLOW | +0.596 |
tten | +0.585 |
�� | +0.571 |
Swordsman | +0.553 |
FACE | +0.548 |
Neg logit contributions | |
ippers | -0.542 |
ipper | -0.530 |
rones | -0.421 |
lems | -0.419 |
mast | -0.411 |
Token | a |
Token ID | 257 |
Feature activation | +7.31e-02 |
Pos logit contributions | |
LLOW | +0.770 |
tten | +0.751 |
�� | +0.736 |
Swordsman | +0.708 |
FACE | +0.701 |
Neg logit contributions | |
ippers | -0.715 |
ipper | -0.702 |
rones | -0.558 |
lems | -0.555 |
lihood | -0.549 |
Token | com |
Token ID | 785 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ippers | +1.884 |
ipper | +1.842 |
rones | +1.482 |
lems | +1.472 |
Carney | +1.462 |
Neg logit contributions | |
LLOW | -1.860 |
tten | -1.813 |
�� | -1.784 |
Swordsman | -1.718 |
ADRA | -1.706 |
Token |
|
Token ID | 198 |
Feature activation | +9.21e-01 |
Pos logit contributions | |
LLOW | +1.146 |
tten | +1.111 |
�� | +1.082 |
Swordsman | +1.049 |
FACE | +1.040 |
Neg logit contributions | |
ippers | -1.213 |
ipper | -1.186 |
lems | -0.961 |
rones | -0.961 |
lihood | -0.951 |
Token |
|
Token ID | 198 |
Feature activation | +7.68e-01 |
Pos logit contributions | |
LLOW | +0.735 |
tten | +0.712 |
�� | +0.687 |
Swordsman | +0.678 |
FACE | +0.674 |
Neg logit contributions | |
ippers | -0.658 |
ipper | -0.640 |
rones | -0.507 |
lems | -0.505 |
lihood | -0.501 |
Token | By |
Token ID | 3886 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
LLOW | +0.517 |
tten | +0.502 |
�� | +0.486 |
FACE | +0.468 |
Swordsman | +0.467 |
Neg logit contributions | |
ippers | -0.643 |
ipper | -0.630 |
rones | -0.519 |
lems | -0.517 |
Carney | -0.516 |
Token | the |
Token ID | 262 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
LLOW | +1.489 |
tten | +1.455 |
�� | +1.408 |
Swordsman | +1.359 |
FACE | +1.341 |
Neg logit contributions | |
ippers | -1.524 |
ipper | -1.485 |
lems | -1.206 |
rones | -1.197 |
lihood | -1.176 |
Token | numbers |
Token ID | 3146 |
Feature activation | +6.84e+00 |
Pos logit contributions | |
LLOW | +1.394 |
tten | +1.332 |
�� | +1.312 |
Swordsman | +1.273 |
FACE | +1.238 |
Neg logit contributions | |
ippers | -1.506 |
ipper | -1.472 |
lems | -1.188 |
rones | -1.183 |
lihood | -1.173 |
Token |
|
Token ID | 198 |
Feature activation | +6.96e-01 |
Pos logit contributions | |
LLOW | +4.954 |
tten | +4.754 |
�� | +4.579 |
Swordsman | +4.498 |
FACE | +4.430 |
Neg logit contributions | |
ippers | -5.182 |
ipper | -5.064 |
lems | -4.151 |
rones | -4.105 |
lihood | -4.063 |
Token |
|
Token ID | 198 |
Feature activation | +5.69e-01 |
Pos logit contributions | |
LLOW | +0.554 |
tten | +0.537 |
�� | +0.519 |
Swordsman | +0.511 |
FACE | +0.509 |
Neg logit contributions | |
ippers | -0.498 |
ipper | -0.485 |
rones | -0.384 |
lems | -0.382 |
lihood | -0.378 |
Token | 12 |
Token ID | 1065 |
Feature activation | +9.52e-01 |
Pos logit contributions | |
LLOW | +0.380 |
tten | +0.369 |
�� | +0.357 |
FACE | +0.345 |
Swordsman | +0.343 |
Neg logit contributions | |
ippers | -0.480 |
ipper | -0.472 |
rones | -0.388 |
lems | -0.387 |
Carney | -0.386 |
Token | , |
Token ID | 11 |
Feature activation | -2.84e-01 |
Pos logit contributions | |
LLOW | +0.719 |
tten | +0.702 |
�� | +0.686 |
Swordsman | +0.667 |
FACE | +0.653 |
Neg logit contributions | |
ippers | -0.719 |
ipper | -0.705 |
rones | -0.565 |
lems | -0.563 |
Carney | -0.558 |
Token | 000 |
Token ID | 830 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
ippers | +0.242 |
ipper | +0.237 |
rones | +0.196 |
lems | +0.196 |
Carney | +0.194 |
Neg logit contributions | |
LLOW | -0.187 |
tten | -0.182 |
�� | -0.177 |
Swordsman | -0.171 |
ADRA | -0.168 |
Token | we |
Token ID | 356 |
Feature activation | -5.92e-03 |
Pos logit contributions | |
LLOW | +0.123 |
tten | +0.121 |
�� | +0.118 |
Swordsman | +0.115 |
FACE | +0.114 |
Neg logit contributions | |
ippers | -0.099 |
ipper | -0.096 |
rones | -0.075 |
lems | -0.075 |
lihood | -0.074 |
Token | � |
Token ID | 447 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
ippers | +0.004 |
ipper | +0.004 |
rones | +0.003 |
lems | +0.003 |
Carney | +0.003 |
Neg logit contributions | |
LLOW | -0.005 |
tten | -0.004 |
�� | -0.004 |
Swordsman | -0.004 |
FACE | -0.004 |
Token | � |
Token ID | 247 |
Feature activation | -1.28e+00 |
Pos logit contributions | |
ippers | +0.768 |
ipper | +0.750 |
rones | +0.600 |
lems | +0.598 |
lihood | +0.594 |
Neg logit contributions | |
LLOW | -0.830 |
tten | -0.806 |
�� | -0.805 |
Swordsman | -0.759 |
ADRA | -0.756 |
Token | ll |
Token ID | 297 |
Feature activation | +1.29e-01 |
Pos logit contributions | |
ippers | +0.916 |
ipper | +0.893 |
rones | +0.710 |
lems | +0.706 |
Carney | +0.701 |
Neg logit contributions | |
LLOW | -1.016 |
tten | -0.996 |
�� | -0.972 |
Swordsman | -0.940 |
FACE | -0.929 |
Token | come |
Token ID | 1282 |
Feature activation | +1.36e-01 |
Pos logit contributions | |
LLOW | +0.105 |
tten | +0.103 |
�� | +0.100 |
Swordsman | +0.098 |
FACE | +0.096 |
Neg logit contributions | |
ippers | -0.089 |
ipper | -0.087 |
rones | -0.069 |
lems | -0.068 |
Carney | -0.068 |
Token | out |
Token ID | 503 |
Feature activation | +6.83e+00 |
Pos logit contributions | |
LLOW | +0.093 |
tten | +0.090 |
�� | +0.087 |
Swordsman | +0.085 |
FACE | +0.083 |
Neg logit contributions | |
ippers | -0.113 |
ipper | -0.110 |
rones | -0.091 |
lems | -0.091 |
Carney | -0.090 |
Token | on |
Token ID | 319 |
Feature activation | +6.37e-01 |
Pos logit contributions | |
LLOW | +4.731 |
tten | +4.634 |
Swordsman | +4.513 |
�� | +4.314 |
Bleach | +4.312 |
Neg logit contributions | |
ippers | -5.199 |
ipper | -5.017 |
rones | -4.127 |
lems | -4.093 |
lihood | -4.076 |
Token | top |
Token ID | 1353 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
LLOW | +0.353 |
tten | +0.346 |
�� | +0.333 |
Swordsman | +0.319 |
FACE | +0.313 |
Neg logit contributions | |
ippers | -0.607 |
ipper | -0.594 |
lems | -0.507 |
rones | -0.503 |
lihood | -0.498 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-01 |
Pos logit contributions | |
ippers | +0.726 |
ipper | +0.709 |
rones | +0.567 |
lems | +0.565 |
Carney | +0.559 |
Neg logit contributions | |
LLOW | -0.760 |
tten | -0.741 |
�� | -0.726 |
Swordsman | -0.702 |
FACE | -0.693 |
Token | � |
Token ID | 447 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
ippers | +0.563 |
ipper | +0.548 |
rones | +0.429 |
lems | +0.428 |
Carney | +0.421 |
Neg logit contributions | |
LLOW | -0.676 |
tten | -0.658 |
�� | -0.646 |
Swordsman | -0.628 |
FACE | -0.621 |
Token | � |
Token ID | 251 |
Feature activation | -6.94e-01 |
Pos logit contributions | |
ippers | +1.268 |
ipper | +1.240 |
rones | +1.003 |
lems | +1.000 |
Carney | +0.991 |
Neg logit contributions | |
LLOW | -1.207 |
tten | -1.175 |
�� | -1.150 |
Swordsman | -1.110 |
ADRA | -1.097 |
Token | her |
Token ID | 607 |
Feature activation | -3.71e-01 |
Pos logit contributions | |
LLOW | +2.237 |
tten | +2.145 |
�� | +2.102 |
Swordsman | +2.053 |
ADRA | +2.018 |
Neg logit contributions | |
ippers | -2.293 |
ipper | -2.219 |
rones | -1.823 |
lems | -1.810 |
mast | -1.773 |
Token | personal |
Token ID | 2614 |
Feature activation | +9.83e-01 |
Pos logit contributions | |
ippers | +0.276 |
ipper | +0.269 |
rones | +0.216 |
lems | +0.215 |
Carney | +0.212 |
Neg logit contributions | |
LLOW | -0.285 |
tten | -0.278 |
�� | -0.272 |
Swordsman | -0.264 |
FACE | -0.260 |
Token | story |
Token ID | 1621 |
Feature activation | -3.31e-01 |
Pos logit contributions | |
LLOW | +0.671 |
tten | +0.654 |
�� | +0.633 |
Swordsman | +0.618 |
FACE | +0.607 |
Neg logit contributions | |
ippers | -0.812 |
ipper | -0.791 |
rones | -0.653 |
lems | -0.650 |
mast | -0.645 |
Token | as |
Token ID | 355 |
Feature activation | -8.50e-01 |
Pos logit contributions | |
ippers | +0.254 |
ipper | +0.248 |
rones | +0.201 |
lems | +0.200 |
Carney | +0.198 |
Neg logit contributions | |
LLOW | -0.246 |
tten | -0.239 |
�� | -0.234 |
Swordsman | -0.226 |
ADRA | -0.224 |
Token | supporting |
Token ID | 6493 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ippers | +0.600 |
ipper | +0.584 |
rones | +0.466 |
lems | +0.463 |
lihood | +0.454 |
Neg logit contributions | |
LLOW | -0.685 |
tten | -0.667 |
�� | -0.654 |
Swordsman | -0.634 |
ADRA | -0.629 |
Token | evidence |
Token ID | 2370 |
Feature activation | -8.35e+00 |
Pos logit contributions | |
ippers | +2.194 |
ipper | +2.133 |
rones | +1.625 |
lems | +1.618 |
Carney | +1.606 |
Neg logit contributions | |
LLOW | -3.075 |
tten | -3.009 |
�� | -2.960 |
Swordsman | -2.872 |
ADRA | -2.847 |
Token | . |
Token ID | 13 |
Feature activation | -6.66e-01 |
Pos logit contributions | |
ippers | +5.384 |
ipper | +5.259 |
Carney | +4.138 |
lems | +4.110 |
lihood | +4.031 |
Neg logit contributions | |
LLOW | -6.939 |
tten | -6.722 |
�� | -6.652 |
ADRA | -6.475 |
Swordsman | -6.307 |
Token |
|
Token ID | 198 |
Feature activation | +1.76e-01 |
Pos logit contributions | |
ippers | +0.487 |
ipper | +0.473 |
rones | +0.381 |
lems | +0.378 |
lihood | +0.370 |
Neg logit contributions | |
LLOW | -0.524 |
tten | -0.505 |
�� | -0.494 |
Swordsman | -0.487 |
FACE | -0.478 |
Token |
|
Token ID | 198 |
Feature activation | -3.81e-01 |
Pos logit contributions | |
LLOW | +0.142 |
tten | +0.138 |
�� | +0.134 |
Swordsman | +0.131 |
FACE | +0.131 |
Neg logit contributions | |
ippers | -0.123 |
ipper | -0.120 |
rones | -0.095 |
lems | -0.094 |
Carney | -0.094 |
Token | Sim |
Token ID | 8890 |
Feature activation | +1.69e-01 |
Pos logit contributions | |
ippers | +0.336 |
ipper | +0.328 |
rones | +0.276 |
lems | +0.274 |
Carney | +0.272 |
Neg logit contributions | |
LLOW | -0.242 |
tten | -0.232 |
�� | -0.223 |
FACE | -0.218 |
Swordsman | -0.215 |
Token | pl |
Token ID | 489 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
LLOW | +0.140 |
tten | +0.138 |
�� | +0.134 |
Swordsman | +0.131 |
FACE | +0.129 |
Neg logit contributions | |
ippers | -0.115 |
ipper | -0.111 |
rones | -0.086 |
lihood | -0.086 |
lems | -0.086 |
Token | org |
Token ID | 2398 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
ippers | +3.382 |
ipper | +3.315 |
lems | +2.667 |
rones | +2.667 |
Carney | +2.637 |
Neg logit contributions | |
LLOW | -3.064 |
tten | -2.975 |
�� | -2.912 |
Swordsman | -2.821 |
ADRA | -2.793 |
Token | / |
Token ID | 14 |
Feature activation | -1.75e-01 |
Pos logit contributions | |
ippers | +1.247 |
ipper | +1.218 |
rones | +0.979 |
lems | +0.974 |
Carney | +0.966 |
Neg logit contributions | |
LLOW | -1.261 |
tten | -1.228 |
�� | -1.206 |
Swordsman | -1.164 |
ADRA | -1.154 |
Token | index |
Token ID | 9630 |
Feature activation | +1.56e+00 |
Pos logit contributions | |
ippers | +0.177 |
ipper | +0.174 |
rones | +0.149 |
lems | +0.149 |
Carney | +0.148 |
Neg logit contributions | |
LLOW | -0.086 |
tten | -0.083 |
�� | -0.080 |
Swordsman | -0.076 |
ADRA | -0.075 |
Token | . |
Token ID | 13 |
Feature activation | -5.36e+00 |
Pos logit contributions | |
LLOW | +1.094 |
tten | +1.078 |
�� | +1.029 |
FACE | +1.003 |
Swordsman | +0.998 |
Neg logit contributions | |
ippers | -1.238 |
ipper | -1.209 |
Carney | -0.992 |
rones | -0.987 |
lems | -0.987 |
Token | php |
Token ID | 10121 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
ippers | +3.296 |
ipper | +3.222 |
rones | +2.390 |
lems | +2.375 |
mast | +2.355 |
Neg logit contributions | |
LLOW | -4.814 |
tten | -4.670 |
�� | -4.641 |
Swordsman | -4.530 |
ADRA | -4.500 |
Token | ? |
Token ID | 30 |
Feature activation | -8.15e+00 |
Pos logit contributions | |
ippers | +1.929 |
ipper | +1.889 |
rones | +1.542 |
lems | +1.534 |
lihood | +1.524 |
Neg logit contributions | |
LLOW | -1.682 |
tten | -1.637 |
�� | -1.608 |
Swordsman | -1.546 |
ADRA | -1.534 |
Token | topic |
Token ID | 26652 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ippers | +5.113 |
ipper | +5.023 |
mast | +3.863 |
service | +3.781 |
lihood | +3.765 |
Neg logit contributions | |
LLOW | -6.914 |
�� | -6.657 |
tten | -6.614 |
Swordsman | -6.610 |
ADRA | -6.577 |
Token | = |
Token ID | 28 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
ippers | +1.262 |
ipper | +1.229 |
rones | +0.953 |
lems | +0.949 |
Carney | +0.939 |
Neg logit contributions | |
LLOW | -1.631 |
tten | -1.595 |
�� | -1.565 |
Swordsman | -1.518 |
ADRA | -1.503 |
Token | 28 |
Token ID | 2078 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ippers | +1.948 |
ipper | +1.910 |
rones | +1.566 |
lems | +1.560 |
Carney | +1.550 |
Neg logit contributions | |
LLOW | -1.612 |
tten | -1.567 |
�� | -1.538 |
Swordsman | -1.473 |
ADRA | -1.458 |
Token | 18 |
Token ID | 1507 |
Feature activation | -2.67e+00 |
Pos logit contributions | |
ippers | +1.726 |
ipper | +1.691 |
rones | +1.394 |
lems | +1.389 |
Carney | +1.379 |
Neg logit contributions | |
LLOW | -1.377 |
tten | -1.338 |
�� | -1.309 |
Swordsman | -1.255 |
ADRA | -1.241 |
Token | 48 |
Token ID | 2780 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ippers | +2.198 |
ipper | +2.158 |
rones | +1.770 |
lems | +1.761 |
Carney | +1.751 |
Neg logit contributions | |
LLOW | -1.828 |
tten | -1.777 |
�� | -1.751 |
Swordsman | -1.667 |
ADRA | -1.657 |
Token | / |
Token ID | 14 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ippers | +1.671 |
ipper | +1.634 |
rones | +1.308 |
lems | +1.299 |
Carney | +1.290 |
Neg logit contributions | |
LLOW | -1.761 |
tten | -1.722 |
�� | -1.687 |
Swordsman | -1.638 |
ADRA | -1.616 |
Token | acs |
Token ID | 16436 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
ippers | +0.905 |
ipper | +0.885 |
rones | +0.712 |
lems | +0.710 |
Carney | +0.703 |
Neg logit contributions | |
LLOW | -0.900 |
tten | -0.877 |
�� | -0.859 |
Swordsman | -0.830 |
ADRA | -0.821 |
Token | 5 |
Token ID | 20 |
Feature activation | -2.01e-01 |
Pos logit contributions | |
LLOW | +0.924 |
tten | +0.903 |
�� | +0.883 |
Swordsman | +0.846 |
FACE | +0.843 |
Neg logit contributions | |
ippers | -0.906 |
ipper | -0.882 |
lems | -0.703 |
rones | -0.703 |
Carney | -0.699 |
Token | / |
Token ID | 14 |
Feature activation | -4.31e-01 |
Pos logit contributions | |
ippers | +0.145 |
ipper | +0.141 |
rones | +0.113 |
lems | +0.112 |
Carney | +0.111 |
Neg logit contributions | |
LLOW | -0.158 |
tten | -0.155 |
�� | -0.151 |
Swordsman | -0.146 |
FACE | -0.145 |
Token | profile |
Token ID | 13317 |
Feature activation | +1.97e+00 |
Pos logit contributions | |
ippers | +0.329 |
ipper | +0.321 |
rones | +0.259 |
lems | +0.258 |
Carney | +0.256 |
Neg logit contributions | |
LLOW | -0.323 |
tten | -0.315 |
�� | -0.308 |
Swordsman | -0.297 |
FACE | -0.294 |
Token | ? |
Token ID | 30 |
Feature activation | -8.01e+00 |
Pos logit contributions | |
LLOW | +1.474 |
tten | +1.437 |
�� | +1.396 |
Swordsman | +1.351 |
FACE | +1.345 |
Neg logit contributions | |
ippers | -1.491 |
ipper | -1.454 |
lems | -1.167 |
rones | -1.164 |
Carney | -1.150 |
Token | get |
Token ID | 1136 |
Feature activation | -1.80e+00 |
Pos logit contributions | |
ippers | +5.995 |
ipper | +5.931 |
mast | +4.742 |
ploy | +4.673 |
rones | +4.649 |
Neg logit contributions | |
LLOW | -5.886 |
�� | -5.687 |
tten | -5.636 |
Swordsman | -5.526 |
ADRA | -5.372 |
Token | = |
Token ID | 28 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
ippers | +1.343 |
ipper | +1.311 |
rones | +1.052 |
lems | +1.049 |
Carney | +1.039 |
Neg logit contributions | |
LLOW | -1.373 |
tten | -1.338 |
�� | -1.310 |
Swordsman | -1.266 |
ADRA | -1.253 |
Token | NAME |
Token ID | 20608 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
ippers | +1.797 |
ipper | +1.762 |
rones | +1.427 |
lems | +1.424 |
mast | +1.413 |
Neg logit contributions | |
LLOW | -1.639 |
tten | -1.595 |
�� | -1.568 |
Swordsman | -1.517 |
ADRA | -1.493 |
Token | , |
Token ID | 11 |
Feature activation | -1.96e+00 |
Pos logit contributions | |
LLOW | +0.998 |
tten | +0.968 |
�� | +0.944 |
Swordsman | +0.923 |
FACE | +0.921 |
Neg logit contributions | |
ippers | -0.847 |
ipper | -0.823 |
lems | -0.647 |
rones | -0.645 |
Carney | -0.629 |
Token | DP |
Token ID | 6322 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ippers | +1.563 |
ipper | +1.532 |
rones | +1.247 |
lems | +1.241 |
Carney | +1.234 |
Neg logit contributions | |
LLOW | -1.397 |
tten | -1.360 |
�� | -1.332 |
Swordsman | -1.285 |
ADRA | -1.266 |
Token | were |
Token ID | 547 |
Feature activation | -2.73e+00 |
Pos logit contributions | |
ippers | +0.307 |
ipper | +0.300 |
rones | +0.238 |
lems | +0.236 |
Carney | +0.232 |
Neg logit contributions | |
LLOW | -0.352 |
tten | -0.345 |
�� | -0.338 |
Swordsman | -0.328 |
FACE | -0.324 |
Token | part |
Token ID | 636 |
Feature activation | -2.87e+00 |
Pos logit contributions | |
ippers | +1.967 |
ipper | +1.916 |
lems | +1.522 |
Carney | +1.520 |
rones | +1.514 |
Neg logit contributions | |
LLOW | -2.160 |
tten | -2.087 |
�� | -2.050 |
Swordsman | -1.993 |
ADRA | -1.967 |
Token | of |
Token ID | 286 |
Feature activation | -3.29e+00 |
Pos logit contributions | |
ippers | +1.990 |
ipper | +1.947 |
rones | +1.523 |
lems | +1.519 |
Carney | +1.514 |
Neg logit contributions | |
LLOW | -2.331 |
tten | -2.271 |
�� | -2.239 |
ADRA | -2.162 |
Swordsman | -2.162 |
Token | the |
Token ID | 262 |
Feature activation | -2.68e+00 |
Pos logit contributions | |
ippers | +2.196 |
ipper | +2.137 |
lems | +1.661 |
Carney | +1.657 |
rones | +1.644 |
Neg logit contributions | |
LLOW | -2.761 |
tten | -2.674 |
�� | -2.631 |
Swordsman | -2.571 |
FACE | -2.551 |
Token | messages |
Token ID | 6218 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ippers | +1.936 |
ipper | +1.890 |
lems | +1.498 |
rones | +1.496 |
Carney | +1.488 |
Neg logit contributions | |
LLOW | -2.114 |
tten | -2.053 |
�� | -2.015 |
Swordsman | -1.951 |
FACE | -1.934 |
Token | Ms |
Token ID | 6997 |
Feature activation | -7.98e+00 |
Pos logit contributions | |
ippers | +1.270 |
ipper | +1.240 |
rones | +0.982 |
lems | +0.981 |
Carney | +0.977 |
Neg logit contributions | |
LLOW | -1.388 |
tten | -1.352 |
�� | -1.328 |
Swordsman | -1.284 |
ADRA | -1.270 |
Token | . |
Token ID | 13 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ippers | +5.058 |
ipper | +4.946 |
Carney | +3.874 |
lems | +3.821 |
rones | +3.740 |
Neg logit contributions | |
LLOW | -6.761 |
�� | -6.754 |
ADRA | -6.617 |
tten | -6.603 |
�� | -6.331 |
Token | Lerner |
Token ID | 49750 |
Feature activation | -5.12e+00 |
Pos logit contributions | |
ippers | +1.895 |
ipper | +1.837 |
Carney | +1.341 |
lems | +1.339 |
rones | +1.327 |
Neg logit contributions | |
LLOW | -3.355 |
tten | -3.290 |
�� | -3.246 |
ADRA | -3.143 |
Swordsman | -3.142 |
Token | lost |
Token ID | 2626 |
Feature activation | -3.14e-02 |
Pos logit contributions | |
ippers | +3.345 |
ipper | +3.270 |
Carney | +2.514 |
rones | +2.498 |
lems | +2.496 |
Neg logit contributions | |
LLOW | -4.382 |
tten | -4.268 |
�� | -4.200 |
ADRA | -4.068 |
Swordsman | -4.054 |
Token | in |
Token ID | 287 |
Feature activation | -1.60e+00 |
Pos logit contributions | |
ippers | +0.023 |
ipper | +0.022 |
lems | +0.018 |
rones | +0.018 |
Carney | +0.017 |
Neg logit contributions | |
LLOW | -0.025 |
tten | -0.024 |
�� | -0.023 |
Swordsman | -0.023 |
FACE | -0.023 |
Token | a |
Token ID | 257 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
ippers | +1.121 |
ipper | +1.094 |
rones | +0.864 |
lems | +0.860 |
Carney | +0.851 |
Neg logit contributions | |
LLOW | -1.290 |
tten | -1.260 |
�� | -1.235 |
Swordsman | -1.196 |
ADRA | -1.184 |
Token | Sen |
Token ID | 2311 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ippers | +0.368 |
ipper | +0.359 |
rones | +0.285 |
lems | +0.284 |
mast | +0.279 |
Neg logit contributions | |
LLOW | -0.404 |
tten | -0.395 |
�� | -0.385 |
Swordsman | -0.375 |
ADRA | -0.369 |
Token | . |
Token ID | 13 |
Feature activation | +6.74e-01 |
Pos logit contributions | |
ippers | +0.932 |
ipper | +0.911 |
rones | +0.732 |
lems | +0.729 |
Carney | +0.723 |
Neg logit contributions | |
LLOW | -0.939 |
tten | -0.915 |
�� | -0.897 |
Swordsman | -0.866 |
ADRA | -0.858 |
Token | Cruz |
Token ID | 8742 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
LLOW | +0.483 |
tten | +0.468 |
�� | +0.458 |
Swordsman | +0.445 |
FACE | +0.441 |
Neg logit contributions | |
ippers | -0.535 |
ipper | -0.523 |
rones | -0.424 |
lems | -0.423 |
lihood | -0.418 |
Token | was |
Token ID | 373 |
Feature activation | -7.11e-01 |
Pos logit contributions | |
ippers | +0.735 |
ipper | +0.718 |
rones | +0.571 |
lems | +0.570 |
Carney | +0.562 |
Neg logit contributions | |
LLOW | -0.785 |
tten | -0.767 |
�� | -0.750 |
Swordsman | -0.727 |
FACE | -0.718 |
Token | no |
Token ID | 645 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
ippers | +0.502 |
ipper | +0.489 |
rones | +0.387 |
lems | +0.385 |
lihood | +0.380 |
Neg logit contributions | |
LLOW | -0.573 |
tten | -0.561 |
�� | -0.546 |
Swordsman | -0.531 |
FACE | -0.528 |
Token | exception |
Token ID | 6631 |
Feature activation | -7.76e+00 |
Pos logit contributions | |
ippers | +1.590 |
ipper | +1.556 |
rones | +1.185 |
Carney | +1.179 |
lems | +1.175 |
Neg logit contributions | |
LLOW | -2.210 |
tten | -2.160 |
�� | -2.128 |
Swordsman | -2.055 |
ADRA | -2.047 |
Token | , |
Token ID | 11 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ippers | +4.963 |
ipper | +4.866 |
Carney | +3.833 |
lems | +3.780 |
rones | +3.759 |
Neg logit contributions | |
LLOW | -6.462 |
tten | -6.268 |
ADRA | -6.220 |
�� | -6.162 |
FACE | -5.972 |
Token | � |
Token ID | 447 |
Feature activation | -2.64e-01 |
Pos logit contributions | |
ippers | +0.925 |
ipper | +0.903 |
rones | +0.724 |
lems | +0.721 |
Carney | +0.715 |
Neg logit contributions | |
LLOW | -0.950 |
tten | -0.926 |
�� | -0.907 |
Swordsman | -0.877 |
ADRA | -0.867 |
Token | � |
Token ID | 251 |
Feature activation | -1.16e+00 |
Pos logit contributions | |
ippers | +0.195 |
ipper | +0.191 |
rones | +0.153 |
lems | +0.152 |
Carney | +0.150 |
Neg logit contributions | |
LLOW | -0.203 |
tten | -0.198 |
�� | -0.195 |
Swordsman | -0.188 |
ADRA | -0.186 |
Token | a |
Token ID | 257 |
Feature activation | -7.43e-01 |
Pos logit contributions | |
ippers | +0.858 |
ipper | +0.838 |
rones | +0.669 |
lems | +0.667 |
Carney | +0.664 |
Neg logit contributions | |
LLOW | -0.900 |
tten | -0.877 |
�� | -0.861 |
Swordsman | -0.831 |
ADRA | -0.822 |
Token | Cruz |
Token ID | 8742 |
Feature activation | -8.59e-01 |
Pos logit contributions | |
ippers | +0.529 |
ipper | +0.516 |
rones | +0.408 |
lems | +0.408 |
Carney | +0.402 |
Neg logit contributions | |
LLOW | -0.593 |
tten | -0.580 |
�� | -0.567 |
Swordsman | -0.550 |
FACE | -0.544 |
Token | our |
Token ID | 674 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
ippers | +0.056 |
ipper | +0.055 |
rones | +0.044 |
lems | +0.044 |
Carney | +0.043 |
Neg logit contributions | |
LLOW | -0.059 |
tten | -0.058 |
�� | -0.056 |
Swordsman | -0.055 |
FACE | -0.054 |
Token | eyebrows |
Token ID | 26928 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
ippers | +1.640 |
ipper | +1.605 |
rones | +1.313 |
lems | +1.309 |
Carney | +1.298 |
Neg logit contributions | |
LLOW | -1.413 |
tten | -1.374 |
�� | -1.344 |
Swordsman | -1.294 |
ADRA | -1.279 |
Token | and |
Token ID | 290 |
Feature activation | -1.27e+00 |
Pos logit contributions | |
ippers | +2.492 |
ipper | +2.432 |
rones | +1.931 |
lems | +1.922 |
Carney | +1.904 |
Neg logit contributions | |
LLOW | -2.767 |
tten | -2.708 |
�� | -2.664 |
Swordsman | -2.571 |
ADRA | -2.560 |
Token | blew |
Token ID | 17948 |
Feature activation | -3.23e+00 |
Pos logit contributions | |
ippers | +0.952 |
ipper | +0.932 |
rones | +0.748 |
lems | +0.747 |
lihood | +0.738 |
Neg logit contributions | |
LLOW | -0.963 |
tten | -0.939 |
�� | -0.913 |
Swordsman | -0.886 |
FACE | -0.879 |
Token | our |
Token ID | 674 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
ippers | +2.174 |
ipper | +2.115 |
rones | +1.653 |
lems | +1.647 |
Carney | +1.632 |
Neg logit contributions | |
LLOW | -2.702 |
tten | -2.639 |
�� | -2.597 |
Swordsman | -2.513 |
ADRA | -2.493 |
Token | minds |
Token ID | 9017 |
Feature activation | -7.38e+00 |
Pos logit contributions | |
ippers | +1.897 |
ipper | +1.852 |
rones | +1.472 |
lems | +1.466 |
Carney | +1.452 |
Neg logit contributions | |
LLOW | -2.074 |
tten | -2.023 |
�� | -1.983 |
Swordsman | -1.919 |
ADRA | -1.898 |
Token | in |
Token ID | 287 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ippers | +4.918 |
ipper | +4.772 |
rones | +3.774 |
Carney | +3.758 |
lihood | +3.746 |
Neg logit contributions | |
LLOW | -5.976 |
tten | -5.886 |
�� | -5.885 |
ADRA | -5.765 |
Swordsman | -5.580 |
Token | 2014 |
Token ID | 1946 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ippers | +1.050 |
ipper | +1.026 |
rones | +0.817 |
lems | +0.816 |
Carney | +0.804 |
Neg logit contributions | |
LLOW | -1.127 |
tten | -1.100 |
�� | -1.076 |
Swordsman | -1.043 |
FACE | -1.029 |
Token | . |
Token ID | 13 |
Feature activation | -3.79e-01 |
Pos logit contributions | |
ippers | +1.536 |
ipper | +1.501 |
rones | +1.204 |
lems | +1.199 |
Carney | +1.188 |
Neg logit contributions | |
LLOW | -1.570 |
tten | -1.531 |
�� | -1.500 |
Swordsman | -1.449 |
ADRA | -1.434 |
Token | Buck |
Token ID | 13452 |
Feature activation | +2.41e-01 |
Pos logit contributions | |
ippers | +0.275 |
ipper | +0.268 |
lems | +0.213 |
rones | +0.213 |
ploy | +0.209 |
Neg logit contributions | |
LLOW | -0.301 |
tten | -0.289 |
�� | -0.283 |
Swordsman | -0.279 |
FACE | -0.275 |
Token | le |
Token ID | 293 |
Feature activation | -5.78e-01 |
Pos logit contributions | |
LLOW | +0.195 |
tten | +0.190 |
�� | +0.186 |
Swordsman | +0.181 |
FACE | +0.179 |
Neg logit contributions | |
ippers | -0.168 |
ipper | -0.165 |
rones | -0.131 |
lems | -0.129 |
lihood | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -3.42e-01 |
Pos logit contributions | |
ippers | +0.372 |
ipper | +0.363 |
rones | +0.291 |
lems | +0.290 |
Carney | +0.287 |
Neg logit contributions | |
LLOW | -0.388 |
tten | -0.378 |
�� | -0.371 |
Swordsman | -0.358 |
ADRA | -0.355 |
Token | It |
Token ID | 632 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ippers | +0.247 |
ipper | +0.242 |
rones | +0.191 |
lems | +0.191 |
lihood | +0.187 |
Neg logit contributions | |
LLOW | -0.271 |
tten | -0.263 |
�� | -0.257 |
Swordsman | -0.251 |
FACE | -0.248 |
Token | was |
Token ID | 373 |
Feature activation | +2.55e-01 |
Pos logit contributions | |
ippers | +1.091 |
ipper | +1.065 |
rones | +0.844 |
lems | +0.842 |
Carney | +0.831 |
Neg logit contributions | |
LLOW | -1.214 |
tten | -1.186 |
�� | -1.161 |
Swordsman | -1.126 |
FACE | -1.111 |
Token | just |
Token ID | 655 |
Feature activation | +5.85e-01 |
Pos logit contributions | |
LLOW | +0.196 |
tten | +0.192 |
�� | +0.187 |
Swordsman | +0.182 |
FACE | +0.180 |
Neg logit contributions | |
ippers | -0.189 |
ipper | -0.184 |
rones | -0.148 |
lems | -0.147 |
ploy | -0.144 |
Token | one |
Token ID | 530 |
Feature activation | +7.21e-01 |
Pos logit contributions | |
LLOW | +0.443 |
tten | +0.432 |
�� | +0.423 |
Swordsman | +0.410 |
FACE | +0.404 |
Neg logit contributions | |
ippers | -0.440 |
ipper | -0.430 |
rones | -0.345 |
lems | -0.345 |
lihood | -0.340 |
Token | example |
Token ID | 1672 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
LLOW | +0.541 |
tten | +0.531 |
�� | +0.513 |
Swordsman | +0.504 |
FACE | +0.493 |
Neg logit contributions | |
ippers | -0.549 |
ipper | -0.534 |
rones | -0.434 |
lems | -0.432 |
mast | -0.420 |
Token | of |
Token ID | 286 |
Feature activation | -3.94e+00 |
Pos logit contributions | |
ippers | +4.465 |
ipper | +4.328 |
Carney | +3.500 |
lihood | +3.303 |
mast | +3.256 |
Neg logit contributions | |
LLOW | -6.265 |
tten | -6.058 |
�� | -6.022 |
ADRA | -5.891 |
Swordsman | -5.796 |
Token | Trump |
Token ID | 1301 |
Feature activation | +8.77e-01 |
Pos logit contributions | |
ippers | +2.783 |
ipper | +2.702 |
Carney | +2.160 |
rones | +2.135 |
lems | +2.124 |
Neg logit contributions | |
LLOW | -3.121 |
tten | -3.053 |
�� | -3.006 |
Swordsman | -2.899 |
ADRA | -2.879 |
Token | attempting |
Token ID | 9361 |
Feature activation | +1.44e-01 |
Pos logit contributions | |
LLOW | +0.664 |
tten | +0.646 |
�� | +0.629 |
Swordsman | +0.610 |
FACE | +0.605 |
Neg logit contributions | |
ippers | -0.658 |
ipper | -0.645 |
rones | -0.520 |
lems | -0.518 |
Carney | -0.511 |
Token | to |
Token ID | 284 |
Feature activation | -1.00e+00 |
Pos logit contributions | |
LLOW | +0.103 |
tten | +0.100 |
�� | +0.098 |
Swordsman | +0.095 |
ADRA | +0.093 |
Neg logit contributions | |
ippers | -0.114 |
ipper | -0.111 |
rones | -0.091 |
lems | -0.090 |
Carney | -0.090 |
Token | mend |
Token ID | 47618 |
Feature activation | -9.88e-01 |
Pos logit contributions | |
ippers | +0.747 |
ipper | +0.730 |
rones | +0.586 |
lems | +0.583 |
Carney | +0.577 |
Neg logit contributions | |
LLOW | -0.766 |
tten | -0.749 |
�� | -0.731 |
Swordsman | -0.707 |
FACE | -0.699 |
Token | appeal |
Token ID | 5198 |
Feature activation | -1.66e+00 |
Pos logit contributions | |
LLOW | +0.405 |
tten | +0.395 |
�� | +0.386 |
Swordsman | +0.373 |
FACE | +0.369 |
Neg logit contributions | |
ippers | -0.416 |
ipper | -0.406 |
rones | -0.329 |
lems | -0.327 |
Carney | -0.324 |
Token | to |
Token ID | 284 |
Feature activation | +8.40e-01 |
Pos logit contributions | |
ippers | +1.231 |
ipper | +1.203 |
rones | +0.963 |
lems | +0.960 |
Carney | +0.951 |
Neg logit contributions | |
LLOW | -1.279 |
tten | -1.247 |
�� | -1.223 |
Swordsman | -1.180 |
ADRA | -1.170 |
Token | remain |
Token ID | 3520 |
Feature activation | -2.34e+00 |
Pos logit contributions | |
LLOW | +0.639 |
tten | +0.628 |
�� | +0.598 |
Swordsman | +0.588 |
FACE | +0.585 |
Neg logit contributions | |
ippers | -0.620 |
ipper | -0.613 |
rones | -0.495 |
lems | -0.491 |
lihood | -0.486 |
Token | somewhat |
Token ID | 6454 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
ippers | +1.725 |
ipper | +1.685 |
rones | +1.352 |
lems | +1.340 |
Carney | +1.326 |
Neg logit contributions | |
LLOW | -1.814 |
tten | -1.769 |
�� | -1.725 |
Swordsman | -1.674 |
FACE | -1.655 |
Token | un |
Token ID | 555 |
Feature activation | -2.79e-01 |
Pos logit contributions | |
ippers | +0.309 |
ipper | +0.302 |
rones | +0.245 |
lems | +0.243 |
Carney | +0.239 |
Neg logit contributions | |
LLOW | -0.305 |
tten | -0.297 |
�� | -0.288 |
Swordsman | -0.280 |
FACE | -0.277 |
Token | spoken |
Token ID | 19842 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
ippers | +0.210 |
ipper | +0.205 |
rones | +0.166 |
lems | +0.165 |
Carney | +0.164 |
Neg logit contributions | |
LLOW | -0.211 |
tten | -0.207 |
�� | -0.201 |
Swordsman | -0.195 |
ADRA | -0.193 |
Token | ? |
Token ID | 30 |
Feature activation | -4.18e+00 |
Pos logit contributions | |
ippers | +5.138 |
ipper | +5.034 |
lems | +3.999 |
Carney | +3.990 |
rones | +3.971 |
Neg logit contributions | |
LLOW | -5.732 |
�� | -5.584 |
tten | -5.583 |
ADRA | -5.359 |
Swordsman | -5.275 |
Token |
|
Token ID | 198 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ippers | +2.741 |
ipper | +2.674 |
rones | +2.076 |
lems | +2.069 |
Carney | +2.068 |
Neg logit contributions | |
LLOW | -3.541 |
tten | -3.480 |
�� | -3.426 |
Swordsman | -3.297 |
ADRA | -3.288 |
Token |
|
Token ID | 198 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ippers | +0.986 |
ipper | +0.961 |
rones | +0.753 |
lems | +0.750 |
Carney | +0.742 |
Neg logit contributions | |
LLOW | -1.186 |
tten | -1.157 |
�� | -1.130 |
Swordsman | -1.100 |
FACE | -1.090 |
Token | Part |
Token ID | 7841 |
Feature activation | +3.76e-01 |
Pos logit contributions | |
ippers | +1.058 |
ipper | +1.034 |
rones | +0.843 |
lems | +0.840 |
Carney | +0.836 |
Neg logit contributions | |
LLOW | -0.926 |
tten | -0.896 |
�� | -0.869 |
FACE | -0.842 |
Swordsman | -0.840 |
Token | of |
Token ID | 286 |
Feature activation | -1.59e+00 |
Pos logit contributions | |
LLOW | +0.290 |
tten | +0.282 |
�� | +0.277 |
Swordsman | +0.268 |
ADRA | +0.264 |
Neg logit contributions | |
ippers | -0.277 |
ipper | -0.270 |
rones | -0.217 |
lems | -0.216 |
Carney | -0.214 |
Token | air |
Token ID | 1633 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
ippers | +1.382 |
ipper | +1.346 |
rones | +1.063 |
lems | +1.057 |
Carney | +1.041 |
Neg logit contributions | |
LLOW | -1.626 |
tten | -1.585 |
�� | -1.550 |
Swordsman | -1.506 |
FACE | -1.490 |
Token | spe |
Token ID | 4125 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ippers | +1.153 |
ipper | +1.111 |
rones | +0.779 |
lems | +0.767 |
Carney | +0.760 |
Neg logit contributions | |
LLOW | -2.423 |
tten | -2.378 |
�� | -2.336 |
Swordsman | -2.279 |
FACE | -2.262 |
Token | eds |
Token ID | 5379 |
Feature activation | -9.38e-01 |
Pos logit contributions | |
ippers | +1.063 |
ipper | +1.044 |
rones | +0.883 |
lems | +0.880 |
Carney | +0.874 |
Neg logit contributions | |
LLOW | -0.629 |
tten | -0.608 |
�� | -0.590 |
Swordsman | -0.563 |
ADRA | -0.554 |
Token | . |
Token ID | 13 |
Feature activation | -2.68e-01 |
Pos logit contributions | |
ippers | +0.713 |
ipper | +0.696 |
rones | +0.563 |
lems | +0.561 |
Carney | +0.554 |
Neg logit contributions | |
LLOW | -0.703 |
tten | -0.684 |
�� | -0.668 |
Swordsman | -0.647 |
FACE | -0.639 |
Token | Un |
Token ID | 791 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
ippers | +0.197 |
ipper | +0.192 |
rones | +0.153 |
lems | +0.153 |
ploy | +0.150 |
Neg logit contributions | |
LLOW | -0.211 |
tten | -0.202 |
�� | -0.198 |
Swordsman | -0.196 |
FACE | -0.193 |
Token | spoken |
Token ID | 19842 |
Feature activation | -7.30e+00 |
Pos logit contributions | |
ippers | +1.227 |
ipper | +1.197 |
rones | +0.946 |
lems | +0.942 |
Carney | +0.933 |
Neg logit contributions | |
LLOW | -1.395 |
tten | -1.361 |
�� | -1.336 |
Swordsman | -1.293 |
ADRA | -1.279 |
Token | , |
Token ID | 11 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ippers | +5.067 |
ipper | +4.963 |
rones | +3.879 |
Carney | +3.860 |
lems | +3.854 |
Neg logit contributions | |
LLOW | -5.812 |
tten | -5.713 |
�� | -5.661 |
ADRA | -5.415 |
Swordsman | -5.394 |
Token | but |
Token ID | 475 |
Feature activation | -2.32e-01 |
Pos logit contributions | |
ippers | +0.813 |
ipper | +0.791 |
rones | +0.636 |
lems | +0.635 |
lihood | +0.623 |
Neg logit contributions | |
LLOW | -0.863 |
tten | -0.836 |
�� | -0.817 |
Swordsman | -0.793 |
FACE | -0.784 |
Token | understood |
Token ID | 7247 |
Feature activation | +9.36e-01 |
Pos logit contributions | |
ippers | +0.175 |
ipper | +0.171 |
lems | +0.139 |
rones | +0.138 |
lihood | +0.135 |
Neg logit contributions | |
LLOW | -0.176 |
tten | -0.169 |
�� | -0.165 |
Swordsman | -0.162 |
FACE | -0.159 |
Token | , |
Token ID | 11 |
Feature activation | +1.86e-01 |
Pos logit contributions | |
LLOW | +0.728 |
tten | +0.705 |
�� | +0.684 |
Swordsman | +0.672 |
FACE | +0.663 |
Neg logit contributions | |
ippers | -0.677 |
ipper | -0.654 |
lems | -0.529 |
ploy | -0.525 |
rones | -0.523 |
Token | was |
Token ID | 373 |
Feature activation | +8.08e-01 |
Pos logit contributions | |
LLOW | +0.149 |
tten | +0.144 |
�� | +0.140 |
Swordsman | +0.136 |
FACE | +0.135 |
Neg logit contributions | |
ippers | -0.133 |
ipper | -0.129 |
lems | -0.103 |
rones | -0.103 |
ploy | -0.103 |
Token | The |
Token ID | 383 |
Feature activation | -1.27e-01 |
Pos logit contributions | |
LLOW | +0.425 |
tten | +0.408 |
Swordsman | +0.397 |
�� | +0.397 |
FACE | +0.387 |
Neg logit contributions | |
ippers | -0.395 |
ipper | -0.386 |
rones | -0.306 |
lems | -0.306 |
lihood | -0.301 |
Token | added |
Token ID | 2087 |
Feature activation | -2.10e+00 |
Pos logit contributions | |
ippers | +0.098 |
ipper | +0.095 |
rones | +0.077 |
lems | +0.077 |
Carney | +0.076 |
Neg logit contributions | |
LLOW | -0.094 |
tten | -0.092 |
�� | -0.089 |
Swordsman | -0.087 |
FACE | -0.085 |
Token | Game |
Token ID | 3776 |
Feature activation | -3.92e-01 |
Pos logit contributions | |
ippers | +1.704 |
ipper | +1.669 |
rones | +1.366 |
lems | +1.360 |
Carney | +1.349 |
Neg logit contributions | |
LLOW | -1.467 |
tten | -1.427 |
�� | -1.396 |
Swordsman | -1.341 |
ADRA | -1.329 |
Token | Pad |
Token ID | 26114 |
Feature activation | -3.61e-01 |
Pos logit contributions | |
ippers | +0.240 |
ipper | +0.233 |
rones | +0.177 |
lems | +0.176 |
Carney | +0.174 |
Neg logit contributions | |
LLOW | -0.352 |
tten | -0.345 |
�� | -0.338 |
Swordsman | -0.329 |
ADRA | -0.326 |
Token | features |
Token ID | 3033 |
Feature activation | +2.03e-01 |
Pos logit contributions | |
ippers | +0.258 |
ipper | +0.250 |
lems | +0.200 |
rones | +0.199 |
Carney | +0.197 |
Neg logit contributions | |
LLOW | -0.288 |
tten | -0.281 |
�� | -0.274 |
Swordsman | -0.268 |
FACE | -0.265 |
Token | elicit |
Token ID | 44055 |
Feature activation | -7.22e+00 |
Pos logit contributions | |
LLOW | +0.157 |
tten | +0.153 |
�� | +0.149 |
Swordsman | +0.145 |
FACE | +0.143 |
Neg logit contributions | |
ippers | -0.150 |
ipper | -0.146 |
rones | -0.117 |
lems | -0.117 |
lihood | -0.115 |
Token | the |
Token ID | 262 |
Feature activation | -4.73e-01 |
Pos logit contributions | |
ippers | +4.927 |
ipper | +4.786 |
rones | +3.786 |
Carney | +3.774 |
lems | +3.755 |
Neg logit contributions | |
LLOW | -5.753 |
tten | -5.669 |
�� | -5.605 |
Swordsman | -5.383 |
ADRA | -5.356 |
Token | same |
Token ID | 976 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
ippers | +0.356 |
ipper | +0.348 |
lems | +0.280 |
rones | +0.280 |
lihood | +0.276 |
Neg logit contributions | |
LLOW | -0.359 |
tten | -0.349 |
�� | -0.339 |
Swordsman | -0.329 |
FACE | -0.326 |
Token | � |
Token ID | 564 |
Feature activation | -1.70e-01 |
Pos logit contributions | |
LLOW | +0.419 |
tten | +0.406 |
�� | +0.394 |
Swordsman | +0.385 |
FACE | +0.382 |
Neg logit contributions | |
ippers | -0.389 |
ipper | -0.379 |
rones | -0.302 |
lems | -0.300 |
mast | -0.298 |
Token | � |
Token ID | 250 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
ippers | +0.121 |
ipper | +0.118 |
rones | +0.094 |
lems | +0.093 |
Carney | +0.092 |
Neg logit contributions | |
LLOW | -0.135 |
tten | -0.132 |
�� | -0.131 |
Swordsman | -0.125 |
ADRA | -0.124 |
Token | You |
Token ID | 1639 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
LLOW | +0.385 |
tten | +0.376 |
�� | +0.364 |
FACE | +0.355 |
Swordsman | +0.351 |
Neg logit contributions | |
ippers | -0.355 |
ipper | -0.346 |
rones | -0.276 |
Carney | -0.274 |
lems | -0.274 |
Token | word |
Token ID | 1573 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
LLOW | +0.531 |
tten | +0.515 |
�� | +0.504 |
Swordsman | +0.493 |
FACE | +0.483 |
Neg logit contributions | |
ippers | -0.492 |
ipper | -0.478 |
rones | -0.383 |
lems | -0.380 |
lihood | -0.378 |
Token | " |
Token ID | 366 |
Feature activation | +5.40e-01 |
Pos logit contributions | |
LLOW | +0.666 |
tten | +0.649 |
�� | +0.635 |
Swordsman | +0.616 |
FACE | +0.613 |
Neg logit contributions | |
ippers | -0.556 |
ipper | -0.540 |
rones | -0.423 |
lems | -0.422 |
Carney | -0.417 |
Token | custom |
Token ID | 23144 |
Feature activation | -8.25e-01 |
Pos logit contributions | |
LLOW | +0.410 |
tten | +0.402 |
�� | +0.390 |
FACE | +0.378 |
Swordsman | +0.376 |
Neg logit contributions | |
ippers | -0.405 |
ipper | -0.393 |
rones | -0.317 |
Carney | -0.315 |
lihood | -0.314 |
Token | ize |
Token ID | 1096 |
Feature activation | +4.32e-02 |
Pos logit contributions | |
ippers | +0.700 |
ipper | +0.685 |
rones | +0.566 |
lems | +0.565 |
Carney | +0.560 |
Neg logit contributions | |
LLOW | -0.546 |
tten | -0.531 |
�� | -0.517 |
Swordsman | -0.498 |
ADRA | -0.491 |
Token | " |
Token ID | 1 |
Feature activation | -8.94e-01 |
Pos logit contributions | |
LLOW | +0.027 |
tten | +0.026 |
�� | +0.025 |
Swordsman | +0.024 |
ADRA | +0.024 |
Neg logit contributions | |
ippers | -0.038 |
ipper | -0.037 |
rones | -0.031 |
lems | -0.031 |
Carney | -0.031 |
Token | disappear |
Token ID | 10921 |
Feature activation | -7.16e+00 |
Pos logit contributions | |
ippers | +0.667 |
ipper | +0.651 |
rones | +0.524 |
lems | +0.520 |
lihood | +0.516 |
Neg logit contributions | |
LLOW | -0.682 |
tten | -0.666 |
�� | -0.649 |
Swordsman | -0.630 |
FACE | -0.623 |
Token | ... |
Token ID | 986 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
ippers | +4.822 |
ipper | +4.735 |
lems | +3.714 |
rones | +3.711 |
Carney | +3.679 |
Neg logit contributions | |
LLOW | -5.824 |
tten | -5.687 |
�� | -5.684 |
ADRA | -5.470 |
Swordsman | -5.375 |
Token | until |
Token ID | 1566 |
Feature activation | -1.87e+00 |
Pos logit contributions | |
ippers | +1.411 |
ipper | +1.377 |
rones | +1.096 |
lems | +1.089 |
Carney | +1.079 |
Neg logit contributions | |
LLOW | -1.544 |
tten | -1.510 |
�� | -1.474 |
Swordsman | -1.429 |
FACE | -1.415 |
Token | I |
Token ID | 314 |
Feature activation | -3.11e-01 |
Pos logit contributions | |
ippers | +1.369 |
ipper | +1.333 |
rones | +1.069 |
lems | +1.067 |
lihood | +1.051 |
Neg logit contributions | |
LLOW | -1.456 |
tten | -1.429 |
�� | -1.392 |
Swordsman | -1.354 |
FACE | -1.332 |
Token | closed |
Token ID | 4838 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
ippers | +0.232 |
ipper | +0.226 |
rones | +0.184 |
lems | +0.183 |
lihood | +0.180 |
Neg logit contributions | |
LLOW | -0.237 |
tten | -0.234 |
�� | -0.224 |
Swordsman | -0.221 |
FACE | -0.216 |
Token | and |
Token ID | 290 |
Feature activation | +9.78e-01 |
Pos logit contributions | |
LLOW | +0.989 |
tten | +0.964 |
�� | +0.942 |
Swordsman | +0.917 |
ADRA | +0.900 |
Neg logit contributions | |
ippers | -0.996 |
ipper | -0.970 |
rones | -0.787 |
lems | -0.782 |
lihood | -0.776 |
Token | outgoing |
Token ID | 28181 |
Feature activation | -4.44e+00 |
Pos logit contributions | |
ippers | +2.289 |
ipper | +2.236 |
Carney | +1.724 |
lems | +1.721 |
rones | +1.712 |
Neg logit contributions | |
LLOW | -2.951 |
tten | -2.891 |
�� | -2.871 |
Swordsman | -2.764 |
ADRA | -2.741 |
Token | and |
Token ID | 290 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ippers | +2.971 |
ipper | +2.907 |
lems | +2.255 |
rones | +2.247 |
Carney | +2.246 |
Neg logit contributions | |
LLOW | -3.704 |
tten | -3.621 |
�� | -3.604 |
ADRA | -3.460 |
Swordsman | -3.458 |
Token | is |
Token ID | 318 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ippers | +0.948 |
ipper | +0.921 |
rones | +0.730 |
lems | +0.724 |
lihood | +0.713 |
Neg logit contributions | |
LLOW | -1.115 |
tten | -1.088 |
�� | -1.058 |
Swordsman | -1.031 |
FACE | -1.028 |
Token | a |
Token ID | 257 |
Feature activation | -9.19e-01 |
Pos logit contributions | |
ippers | +0.894 |
ipper | +0.869 |
rones | +0.698 |
lems | +0.692 |
lihood | +0.678 |
Neg logit contributions | |
LLOW | -0.960 |
tten | -0.938 |
�� | -0.915 |
Swordsman | -0.889 |
FACE | -0.883 |
Token | good |
Token ID | 922 |
Feature activation | -5.66e-01 |
Pos logit contributions | |
ippers | +0.687 |
ipper | +0.663 |
lems | +0.541 |
rones | +0.536 |
mast | +0.524 |
Neg logit contributions | |
LLOW | -0.697 |
tten | -0.688 |
�� | -0.659 |
Swordsman | -0.650 |
FACE | -0.648 |
Token | conversation |
Token ID | 5273 |
Feature activation | -7.07e+00 |
Pos logit contributions | |
ippers | +0.411 |
ipper | +0.398 |
lems | +0.320 |
rones | +0.319 |
ploy | +0.313 |
Neg logit contributions | |
LLOW | -0.442 |
tten | -0.432 |
�� | -0.419 |
Swordsman | -0.413 |
FACE | -0.411 |
Token | starter |
Token ID | 14217 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ipper | +4.195 |
ippers | +4.164 |
mast | +3.057 |
Carney | +3.035 |
rones | +3.019 |
Neg logit contributions | |
LLOW | -6.348 |
tten | -6.182 |
�� | -6.088 |
Swordsman | -5.874 |
ADRA | -5.821 |
Token | . |
Token ID | 13 |
Feature activation | -8.42e-02 |
Pos logit contributions | |
ippers | +1.552 |
ipper | +1.516 |
rones | +1.207 |
lems | +1.203 |
Carney | +1.192 |
Neg logit contributions | |
LLOW | -1.669 |
tten | -1.629 |
�� | -1.596 |
Swordsman | -1.544 |
ADRA | -1.527 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
ippers | +0.060 |
ipper | +0.059 |
rones | +0.046 |
lems | +0.046 |
lihood | +0.045 |
Neg logit contributions | |
LLOW | -0.068 |
tten | -0.065 |
�� | -0.064 |
Swordsman | -0.063 |
FACE | -0.062 |
Token | Well |
Token ID | 5779 |
Feature activation | +1.48e-01 |
Pos logit contributions | |
LLOW | +1.585 |
tten | +1.520 |
�� | +1.465 |
FACE | +1.447 |
ATH | +1.418 |
Neg logit contributions | |
ippers | -2.008 |
ipper | -1.968 |
Carney | -1.634 |
rones | -1.629 |
lems | -1.621 |
Token | , |
Token ID | 11 |
Feature activation | +8.89e-02 |
Pos logit contributions | |
LLOW | +0.116 |
tten | +0.113 |
�� | +0.111 |
Swordsman | +0.107 |
FACE | +0.106 |
Neg logit contributions | |
ippers | -0.108 |
ipper | -0.105 |
rones | -0.084 |
lems | -0.084 |
Carney | -0.083 |
Token | is |
Token ID | 318 |
Feature activation | -5.29e-01 |
Pos logit contributions | |
ippers | +2.184 |
ipper | +2.128 |
rones | +1.655 |
lems | +1.648 |
Carney | +1.632 |
Neg logit contributions | |
LLOW | -2.767 |
tten | -2.703 |
�� | -2.654 |
Swordsman | -2.574 |
ADRA | -2.549 |
Token | just |
Token ID | 655 |
Feature activation | +7.27e-02 |
Pos logit contributions | |
ippers | +0.382 |
ipper | +0.373 |
rones | +0.297 |
lems | +0.295 |
mast | +0.290 |
Neg logit contributions | |
LLOW | -0.417 |
tten | -0.409 |
�� | -0.400 |
Swordsman | -0.386 |
FACE | -0.383 |
Token | one |
Token ID | 530 |
Feature activation | +4.43e-01 |
Pos logit contributions | |
LLOW | +0.058 |
tten | +0.057 |
�� | +0.056 |
Swordsman | +0.054 |
ADRA | +0.053 |
Neg logit contributions | |
ippers | -0.052 |
ipper | -0.050 |
rones | -0.040 |
lems | -0.040 |
Carney | -0.039 |
Token | of |
Token ID | 286 |
Feature activation | -5.02e+00 |
Pos logit contributions | |
LLOW | +0.332 |
tten | +0.326 |
�� | +0.315 |
Swordsman | +0.306 |
FACE | +0.303 |
Neg logit contributions | |
ippers | -0.339 |
ipper | -0.330 |
rones | -0.268 |
lems | -0.267 |
mast | -0.260 |
Token | many |
Token ID | 867 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
ippers | +3.830 |
ipper | +3.727 |
Carney | +3.053 |
lihood | +3.031 |
lems | +3.013 |
Neg logit contributions | |
LLOW | -3.676 |
tten | -3.557 |
�� | -3.498 |
Swordsman | -3.376 |
ADRA | -3.351 |
Token | instances |
Token ID | 10245 |
Feature activation | -7.05e+00 |
Pos logit contributions | |
ippers | +1.293 |
ipper | +1.263 |
rones | +1.008 |
lems | +1.005 |
Carney | +0.997 |
Neg logit contributions | |
LLOW | -1.367 |
tten | -1.333 |
�� | -1.307 |
Swordsman | -1.263 |
ADRA | -1.250 |
Token | of |
Token ID | 286 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
ippers | +4.547 |
ipper | +4.369 |
Carney | +3.553 |
lihood | +3.364 |
rones | +3.354 |
Neg logit contributions | |
LLOW | -5.892 |
tten | -5.737 |
�� | -5.651 |
FACE | -5.401 |
ADRA | -5.384 |
Token | legal |
Token ID | 2742 |
Feature activation | +1.20e-01 |
Pos logit contributions | |
ippers | +1.312 |
ipper | +1.282 |
rones | +1.026 |
lems | +1.023 |
Carney | +1.013 |
Neg logit contributions | |
LLOW | -1.359 |
tten | -1.325 |
�� | -1.298 |
Swordsman | -1.255 |
ADRA | -1.241 |
Token | American |
Token ID | 1605 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
LLOW | +0.091 |
tten | +0.089 |
�� | +0.086 |
Swordsman | +0.084 |
ADRA | +0.083 |
Neg logit contributions | |
ippers | -0.090 |
ipper | -0.088 |
rones | -0.071 |
lems | -0.071 |
lihood | -0.071 |
Token | citizens |
Token ID | 4290 |
Feature activation | +6.92e-01 |
Pos logit contributions | |
LLOW | +1.165 |
tten | +1.145 |
�� | +1.104 |
Swordsman | +1.085 |
ADRA | +1.083 |
Neg logit contributions | |
ippers | -1.218 |
ipper | -1.195 |
lems | -0.959 |
rones | -0.958 |
lihood | -0.953 |
Token | being |
Token ID | 852 |
Feature activation | -2.87e-01 |
Pos logit contributions | |
LLOW | +0.476 |
tten | +0.464 |
�� | +0.449 |
Swordsman | +0.436 |
FACE | +0.431 |
Neg logit contributions | |
ippers | -0.565 |
ipper | -0.554 |
rones | -0.458 |
lems | -0.456 |
Carney | -0.448 |
Token | being |
Token ID | 852 |
Feature activation | -1.49e+00 |
Pos logit contributions | |
ippers | +0.235 |
ipper | +0.230 |
rones | +0.185 |
lems | +0.184 |
Carney | +0.182 |
Neg logit contributions | |
LLOW | -0.237 |
tten | -0.231 |
�� | -0.226 |
Swordsman | -0.218 |
FACE | -0.216 |
Token | seen |
Token ID | 1775 |
Feature activation | +1.00e+00 |
Pos logit contributions | |
ippers | +1.218 |
ipper | +1.192 |
rones | +0.973 |
lems | +0.972 |
Carney | +0.970 |
Neg logit contributions | |
LLOW | -1.022 |
tten | -0.994 |
�� | -0.971 |
Swordsman | -0.938 |
ADRA | -0.929 |
Token | as |
Token ID | 355 |
Feature activation | -6.54e-01 |
Pos logit contributions | |
LLOW | +0.735 |
tten | +0.716 |
�� | +0.702 |
Swordsman | +0.676 |
FACE | +0.669 |
Neg logit contributions | |
ippers | -0.778 |
ipper | -0.760 |
rones | -0.615 |
lems | -0.614 |
Carney | -0.607 |
Token | the |
Token ID | 262 |
Feature activation | -9.15e-01 |
Pos logit contributions | |
ippers | +0.469 |
ipper | +0.457 |
rones | +0.365 |
lems | +0.362 |
lihood | +0.357 |
Neg logit contributions | |
LLOW | -0.520 |
tten | -0.508 |
�� | -0.498 |
Swordsman | -0.481 |
ADRA | -0.477 |
Token | latest |
Token ID | 3452 |
Feature activation | -2.07e+00 |
Pos logit contributions | |
ippers | +0.772 |
ipper | +0.755 |
rones | +0.626 |
lems | +0.625 |
lihood | +0.616 |
Neg logit contributions | |
LLOW | -0.610 |
tten | -0.593 |
�� | -0.578 |
Swordsman | -0.556 |
FACE | -0.548 |
Token | instance |
Token ID | 4554 |
Feature activation | -7.02e+00 |
Pos logit contributions | |
ippers | +1.449 |
ipper | +1.415 |
rones | +1.115 |
lems | +1.111 |
Carney | +1.102 |
Neg logit contributions | |
LLOW | -1.676 |
tten | -1.635 |
�� | -1.605 |
Swordsman | -1.554 |
ADRA | -1.538 |
Token | of |
Token ID | 286 |
Feature activation | -2.76e+00 |
Pos logit contributions | |
ippers | +4.305 |
ipper | +4.183 |
Carney | +3.329 |
ploy | +3.117 |
rones | +3.097 |
Neg logit contributions | |
LLOW | -6.023 |
tten | -5.835 |
�� | -5.766 |
Swordsman | -5.641 |
ADRA | -5.598 |
Token | a |
Token ID | 257 |
Feature activation | -8.73e-01 |
Pos logit contributions | |
ippers | +2.050 |
ipper | +2.002 |
rones | +1.598 |
lems | +1.594 |
Carney | +1.592 |
Neg logit contributions | |
LLOW | -2.111 |
tten | -2.056 |
�� | -2.023 |
Swordsman | -1.950 |
ADRA | -1.933 |
Token | new |
Token ID | 649 |
Feature activation | -6.70e-01 |
Pos logit contributions | |
ippers | +0.660 |
ipper | +0.645 |
rones | +0.520 |
lems | +0.520 |
lihood | +0.511 |
Neg logit contributions | |
LLOW | -0.659 |
tten | -0.644 |
�� | -0.626 |
Swordsman | -0.608 |
FACE | -0.602 |
Token | mood |
Token ID | 10038 |
Feature activation | -1.54e+00 |
Pos logit contributions | |
ippers | +0.522 |
ipper | +0.508 |
rones | +0.413 |
lems | +0.411 |
Carney | +0.403 |
Neg logit contributions | |
LLOW | -0.494 |
tten | -0.482 |
�� | -0.467 |
Swordsman | -0.455 |
FACE | -0.451 |
Token | of |
Token ID | 286 |
Feature activation | +5.90e-02 |
Pos logit contributions | |
ippers | +1.073 |
ipper | +1.047 |
rones | +0.824 |
lems | +0.821 |
Carney | +0.814 |
Neg logit contributions | |
LLOW | -1.250 |
tten | -1.220 |
�� | -1.198 |
Swordsman | -1.160 |
ADRA | -1.149 |
Token | com |
Token ID | 785 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
ippers | +3.288 |
ipper | +3.237 |
lems | +2.534 |
mast | +2.529 |
rones | +2.515 |
Neg logit contributions | |
LLOW | -3.895 |
�� | -3.884 |
tten | -3.811 |
ADRA | -3.775 |
Swordsman | -3.665 |
Token | / |
Token ID | 14 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
ippers | +1.440 |
ipper | +1.407 |
rones | +1.131 |
lems | +1.127 |
Carney | +1.116 |
Neg logit contributions | |
LLOW | -1.453 |
tten | -1.418 |
�� | -1.393 |
Swordsman | -1.342 |
ADRA | -1.333 |
Token | iv |
Token ID | 452 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
ippers | +1.630 |
ipper | +1.593 |
rones | +1.312 |
lems | +1.308 |
mast | +1.290 |
Neg logit contributions | |
LLOW | -1.374 |
tten | -1.344 |
�� | -1.316 |
Swordsman | -1.279 |
ADRA | -1.256 |
Token | xx |
Token ID | 5324 |
Feature activation | -2.86e-01 |
Pos logit contributions | |
LLOW | +0.763 |
tten | +0.744 |
�� | +0.723 |
Swordsman | +0.694 |
FACE | +0.687 |
Neg logit contributions | |
ippers | -0.919 |
ipper | -0.900 |
rones | -0.738 |
lems | -0.737 |
Carney | -0.736 |
Token | brand |
Token ID | 17938 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
ippers | +0.210 |
ipper | +0.205 |
rones | +0.164 |
lems | +0.163 |
Carney | +0.162 |
Neg logit contributions | |
LLOW | -0.222 |
tten | -0.217 |
�� | -0.212 |
Swordsman | -0.205 |
FACE | -0.203 |
Token | ? |
Token ID | 30 |
Feature activation | -7.01e+00 |
Pos logit contributions | |
ippers | +0.868 |
ipper | +0.847 |
rones | +0.674 |
lems | +0.671 |
Carney | +0.665 |
Neg logit contributions | |
LLOW | -0.940 |
tten | -0.917 |
�� | -0.898 |
Swordsman | -0.869 |
FACE | -0.860 |
Token | f |
Token ID | 69 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ippers | +4.926 |
ipper | +4.875 |
mast | +3.849 |
lems | +3.763 |
Carney | +3.704 |
Neg logit contributions | |
LLOW | -5.409 |
�� | -5.269 |
tten | -5.256 |
Swordsman | -5.142 |
ADRA | -4.998 |
Token | ref |
Token ID | 5420 |
Feature activation | -3.04e+00 |
Pos logit contributions | |
ippers | +0.807 |
ipper | +0.791 |
rones | +0.637 |
lems | +0.634 |
Carney | +0.630 |
Neg logit contributions | |
LLOW | -0.759 |
tten | -0.740 |
�� | -0.727 |
Swordsman | -0.703 |
ADRA | -0.692 |
Token | = |
Token ID | 28 |
Feature activation | +3.86e-01 |
Pos logit contributions | |
ippers | +1.969 |
ipper | +1.917 |
rones | +1.481 |
lems | +1.470 |
Carney | +1.455 |
Neg logit contributions | |
LLOW | -2.612 |
tten | -2.555 |
�� | -2.519 |
Swordsman | -2.444 |
ADRA | -2.417 |
Token | ts |
Token ID | 912 |
Feature activation | -2.36e-01 |
Pos logit contributions | |
LLOW | +0.372 |
tten | +0.366 |
�� | +0.358 |
Swordsman | +0.349 |
FACE | +0.347 |
Neg logit contributions | |
ippers | -0.211 |
ipper | -0.204 |
rones | -0.149 |
lems | -0.147 |
Carney | -0.145 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
ippers | +0.172 |
ipper | +0.168 |
rones | +0.134 |
lems | +0.133 |
Carney | +0.132 |
Neg logit contributions | |
LLOW | -0.185 |
tten | -0.180 |
�� | -0.177 |
Swordsman | -0.171 |
FACE | -0.170 |
Token | access |
Token ID | 1895 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
ippers | +0.981 |
ipper | +0.958 |
rones | +0.767 |
lems | +0.763 |
Carney | +0.757 |
Neg logit contributions | |
LLOW | -1.026 |
tten | -1.001 |
�� | -0.981 |
Swordsman | -0.948 |
ADRA | -0.938 |
Token | to |
Token ID | 284 |
Feature activation | +5.88e-01 |
Pos logit contributions | |
LLOW | +0.623 |
tten | +0.608 |
�� | +0.594 |
Swordsman | +0.572 |
ADRA | +0.565 |
Neg logit contributions | |
ippers | -0.712 |
ipper | -0.697 |
rones | -0.569 |
lems | -0.568 |
Carney | -0.563 |
Token | calendars |
Token ID | 50215 |
Feature activation | +5.14e-01 |
Pos logit contributions | |
LLOW | +0.454 |
tten | +0.445 |
�� | +0.432 |
Swordsman | +0.421 |
FACE | +0.417 |
Neg logit contributions | |
ippers | -0.432 |
ipper | -0.425 |
lems | -0.342 |
rones | -0.336 |
lihood | -0.331 |
Token | . |
Token ID | 13 |
Feature activation | -6.39e-01 |
Pos logit contributions | |
LLOW | +0.401 |
tten | +0.391 |
�� | +0.382 |
Swordsman | +0.371 |
FACE | +0.367 |
Neg logit contributions | |
ippers | -0.374 |
ipper | -0.366 |
lems | -0.291 |
rones | -0.290 |
Carney | -0.287 |
Token | One |
Token ID | 1881 |
Feature activation | -1.03e+00 |
Pos logit contributions | |
ippers | +0.454 |
ipper | +0.443 |
rones | +0.350 |
lems | +0.349 |
lihood | +0.342 |
Neg logit contributions | |
LLOW | -0.513 |
tten | -0.497 |
�� | -0.487 |
Swordsman | -0.475 |
FACE | -0.470 |
Token | cabinet |
Token ID | 13447 |
Feature activation | -6.97e+00 |
Pos logit contributions | |
ippers | +0.767 |
ipper | +0.749 |
lems | +0.603 |
rones | +0.600 |
Carney | +0.589 |
Neg logit contributions | |
LLOW | -0.791 |
tten | -0.771 |
�� | -0.753 |
Swordsman | -0.732 |
FACE | -0.722 |
Token | pick |
Token ID | 2298 |
Feature activation | -8.63e-01 |
Pos logit contributions | |
ippers | +4.529 |
ipper | +4.453 |
rones | +3.460 |
Carney | +3.447 |
mast | +3.345 |
Neg logit contributions | |
LLOW | -5.823 |
tten | -5.719 |
�� | -5.592 |
Swordsman | -5.429 |
FACE | -5.379 |
Token | stands |
Token ID | 6296 |
Feature activation | -4.96e-01 |
Pos logit contributions | |
ippers | +0.623 |
ipper | +0.608 |
rones | +0.484 |
lems | +0.482 |
Carney | +0.476 |
Neg logit contributions | |
LLOW | -0.681 |
tten | -0.664 |
�� | -0.650 |
Swordsman | -0.631 |
FACE | -0.621 |
Token | out |
Token ID | 503 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
ippers | +0.423 |
ipper | +0.414 |
rones | +0.344 |
lems | +0.342 |
Carney | +0.338 |
Neg logit contributions | |
LLOW | -0.325 |
tten | -0.315 |
�� | -0.305 |
Swordsman | -0.296 |
FACE | -0.291 |
Token | , |
Token ID | 11 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
LLOW | +1.968 |
tten | +1.918 |
�� | +1.857 |
Swordsman | +1.826 |
FACE | +1.774 |
Neg logit contributions | |
ippers | -2.046 |
ipper | -2.006 |
rones | -1.619 |
lems | -1.609 |
ploy | -1.592 |
Token | however |
Token ID | 2158 |
Feature activation | -1.44e+00 |
Pos logit contributions | |
ippers | +0.760 |
ipper | +0.743 |
rones | +0.595 |
lems | +0.593 |
lihood | +0.582 |
Neg logit contributions | |
LLOW | -0.811 |
tten | -0.790 |
�� | -0.770 |
Swordsman | -0.752 |
FACE | -0.739 |
Token | to |
Token ID | 284 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
ippers | +0.715 |
ipper | +0.698 |
rones | +0.557 |
lems | +0.554 |
Carney | +0.549 |
Neg logit contributions | |
LLOW | -0.770 |
tten | -0.751 |
�� | -0.736 |
Swordsman | -0.712 |
ADRA | -0.704 |
Token | discover |
Token ID | 7073 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
ippers | +1.041 |
ipper | +1.015 |
rones | +0.795 |
lems | +0.792 |
Carney | +0.781 |
Neg logit contributions | |
LLOW | -1.264 |
tten | -1.237 |
�� | -1.209 |
Swordsman | -1.175 |
ADRA | -1.162 |
Token | the |
Token ID | 262 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ippers | +1.555 |
ipper | +1.519 |
rones | +1.208 |
lems | +1.206 |
Carney | +1.197 |
Neg logit contributions | |
LLOW | -1.664 |
tten | -1.626 |
�� | -1.596 |
Swordsman | -1.541 |
ADRA | -1.524 |
Token | most |
Token ID | 749 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
ippers | +0.927 |
ipper | +0.906 |
rones | +0.726 |
lems | +0.724 |
Carney | +0.718 |
Neg logit contributions | |
LLOW | -0.946 |
tten | -0.922 |
�� | -0.903 |
Swordsman | -0.873 |
ADRA | -0.863 |
Token | used |
Token ID | 973 |
Feature activation | +8.68e-01 |
Pos logit contributions | |
ippers | +1.078 |
ipper | +1.056 |
lems | +0.877 |
rones | +0.877 |
Carney | +0.873 |
Neg logit contributions | |
LLOW | -0.778 |
tten | -0.755 |
�� | -0.738 |
Swordsman | -0.707 |
ADRA | -0.698 |
Token | Clojure |
Token ID | 43848 |
Feature activation | -6.88e+00 |
Pos logit contributions | |
LLOW | +0.664 |
tten | +0.647 |
�� | +0.627 |
Swordsman | +0.612 |
FACE | +0.603 |
Neg logit contributions | |
ippers | -0.647 |
ipper | -0.631 |
rones | -0.508 |
lems | -0.505 |
Carney | -0.498 |
Token | testing |
Token ID | 4856 |
Feature activation | -2.58e+00 |
Pos logit contributions | |
ippers | +4.674 |
ipper | +4.567 |
rones | +3.558 |
lems | +3.549 |
ploy | +3.436 |
Neg logit contributions | |
LLOW | -5.640 |
tten | -5.536 |
�� | -5.467 |
Swordsman | -5.266 |
FACE | -5.214 |
Token | library |
Token ID | 5888 |
Feature activation | -1.20e-01 |
Pos logit contributions | |
ippers | +1.749 |
ipper | +1.706 |
rones | +1.329 |
lems | +1.327 |
Carney | +1.312 |
Neg logit contributions | |
LLOW | -2.142 |
tten | -2.097 |
�� | -2.068 |
Swordsman | -1.990 |
ADRA | -1.978 |
Token | . |
Token ID | 13 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ippers | +0.089 |
ipper | +0.087 |
rones | +0.070 |
lems | +0.069 |
Carney | +0.069 |
Neg logit contributions | |
LLOW | -0.092 |
tten | -0.090 |
�� | -0.087 |
Swordsman | -0.085 |
FACE | -0.084 |
Token | The |
Token ID | 383 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
ippers | +0.884 |
ipper | +0.862 |
rones | +0.680 |
lems | +0.677 |
Carney | +0.669 |
Neg logit contributions | |
LLOW | -1.022 |
tten | -0.996 |
�� | -0.976 |
Swordsman | -0.948 |
ADRA | -0.936 |
Token | answer |
Token ID | 3280 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
ippers | +1.077 |
ipper | +1.052 |
rones | +0.841 |
lems | +0.838 |
Carney | +0.830 |
Neg logit contributions | |
LLOW | -1.133 |
tten | -1.105 |
�� | -1.083 |
Swordsman | -1.047 |
ADRA | -1.035 |
Token | youtube |
Token ID | 11604 |
Feature activation | -1.10e+00 |
Pos logit contributions | |
ippers | +2.394 |
ipper | +2.349 |
rones | +1.958 |
lems | +1.952 |
mast | +1.934 |
Neg logit contributions | |
LLOW | -1.711 |
tten | -1.638 |
�� | -1.621 |
Swordsman | -1.553 |
ADRA | -1.542 |
Token | . |
Token ID | 13 |
Feature activation | -4.00e+00 |
Pos logit contributions | |
ippers | +0.836 |
ipper | +0.817 |
rones | +0.658 |
lems | +0.655 |
Carney | +0.649 |
Neg logit contributions | |
LLOW | -0.828 |
tten | -0.807 |
�� | -0.788 |
Swordsman | -0.763 |
FACE | -0.754 |
Token | com |
Token ID | 785 |
Feature activation | -5.43e-01 |
Pos logit contributions | |
ippers | +2.898 |
ipper | +2.845 |
rones | +2.253 |
lems | +2.242 |
mast | +2.230 |
Neg logit contributions | |
LLOW | -3.137 |
�� | -3.056 |
tten | -3.043 |
ADRA | -2.958 |
Swordsman | -2.915 |
Token | / |
Token ID | 14 |
Feature activation | -7.78e-01 |
Pos logit contributions | |
ippers | +0.415 |
ipper | +0.405 |
rones | +0.327 |
lems | +0.326 |
Carney | +0.323 |
Neg logit contributions | |
LLOW | -0.406 |
tten | -0.395 |
�� | -0.386 |
Swordsman | -0.373 |
FACE | -0.369 |
Token | watch |
Token ID | 8340 |
Feature activation | -3.77e-01 |
Pos logit contributions | |
ippers | +0.561 |
ipper | +0.548 |
rones | +0.435 |
lems | +0.434 |
Carney | +0.427 |
Neg logit contributions | |
LLOW | -0.614 |
tten | -0.598 |
�� | -0.588 |
Swordsman | -0.571 |
ADRA | -0.562 |
Token | ? |
Token ID | 30 |
Feature activation | -6.81e+00 |
Pos logit contributions | |
ippers | +0.308 |
ipper | +0.301 |
rones | +0.247 |
lems | +0.246 |
Carney | +0.244 |
Neg logit contributions | |
LLOW | -0.262 |
tten | -0.255 |
�� | -0.250 |
Swordsman | -0.240 |
ADRA | -0.238 |
Token | v |
Token ID | 85 |
Feature activation | -7.55e-01 |
Pos logit contributions | |
ippers | +4.048 |
ipper | +3.962 |
mast | +2.909 |
rones | +2.893 |
lems | +2.875 |
Neg logit contributions | |
LLOW | -6.120 |
�� | -5.929 |
tten | -5.858 |
ADRA | -5.820 |
Swordsman | -5.783 |
Token | = |
Token ID | 28 |
Feature activation | +6.94e-01 |
Pos logit contributions | |
ippers | +0.515 |
ipper | +0.502 |
lems | +0.393 |
rones | +0.392 |
Carney | +0.391 |
Neg logit contributions | |
LLOW | -0.624 |
tten | -0.612 |
�� | -0.594 |
Swordsman | -0.577 |
FACE | -0.571 |
Token | 2 |
Token ID | 17 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
LLOW | +0.507 |
tten | +0.494 |
�� | +0.477 |
FACE | +0.461 |
Swordsman | +0.460 |
Neg logit contributions | |
ippers | -0.541 |
ipper | -0.528 |
rones | -0.430 |
Carney | -0.429 |
lihood | -0.428 |
Token | g |
Token ID | 70 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
LLOW | +0.189 |
tten | +0.184 |
�� | +0.178 |
FACE | +0.172 |
Swordsman | +0.172 |
Neg logit contributions | |
ippers | -0.204 |
ipper | -0.199 |
rones | -0.162 |
Carney | -0.162 |
lems | -0.161 |
Token | 7 |
Token ID | 22 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
ippers | +0.760 |
ipper | +0.744 |
rones | +0.610 |
lems | +0.607 |
Carney | +0.604 |
Neg logit contributions | |
LLOW | -0.649 |
tten | -0.631 |
�� | -0.615 |
Swordsman | -0.593 |
ADRA | -0.586 |
Token | category |
Token ID | 6536 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
ippers | +0.449 |
ipper | +0.439 |
rones | +0.351 |
lems | +0.350 |
Carney | +0.346 |
Neg logit contributions | |
LLOW | -0.468 |
tten | -0.456 |
�� | -0.447 |
Swordsman | -0.432 |
ADRA | -0.427 |
Token | , |
Token ID | 11 |
Feature activation | -9.52e-01 |
Pos logit contributions | |
ippers | +2.378 |
ipper | +2.320 |
rones | +1.819 |
Carney | +1.800 |
lems | +1.794 |
Neg logit contributions | |
LLOW | -2.908 |
tten | -2.838 |
�� | -2.786 |
Swordsman | -2.702 |
ADRA | -2.693 |
Token | where |
Token ID | 810 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
ippers | +0.666 |
ipper | +0.649 |
rones | +0.512 |
lems | +0.510 |
Carney | +0.504 |
Neg logit contributions | |
LLOW | -0.772 |
tten | -0.754 |
�� | -0.739 |
Swordsman | -0.716 |
ADRA | -0.708 |
Token | Emmy |
Token ID | 33953 |
Feature activation | +6.24e-01 |
Pos logit contributions | |
ippers | +1.558 |
ipper | +1.523 |
rones | +1.226 |
lems | +1.218 |
Carney | +1.214 |
Neg logit contributions | |
LLOW | -1.550 |
tten | -1.512 |
�� | -1.482 |
Swordsman | -1.429 |
ADRA | -1.418 |
Token | - |
Token ID | 12 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
LLOW | +0.477 |
tten | +0.468 |
�� | +0.455 |
Swordsman | +0.442 |
ADRA | +0.439 |
Neg logit contributions | |
ippers | -0.459 |
ipper | -0.451 |
lems | -0.357 |
rones | -0.356 |
mast | -0.355 |
Token | neg |
Token ID | 12480 |
Feature activation | -6.77e+00 |
Pos logit contributions | |
ippers | +0.989 |
ipper | +0.962 |
rones | +0.723 |
lems | +0.721 |
Carney | +0.715 |
Neg logit contributions | |
LLOW | -1.490 |
tten | -1.460 |
�� | -1.433 |
Swordsman | -1.391 |
FACE | -1.381 |
Token | lected |
Token ID | 12609 |
Feature activation | -3.15e+00 |
Pos logit contributions | |
ippers | +4.691 |
ipper | +4.553 |
rones | +3.588 |
ploy | +3.496 |
lems | +3.487 |
Neg logit contributions | |
LLOW | -5.524 |
tten | -5.295 |
�� | -5.281 |
FACE | -5.230 |
Swordsman | -5.166 |
Token | The |
Token ID | 383 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ippers | +2.395 |
ipper | +2.345 |
rones | +1.894 |
lems | +1.877 |
Carney | +1.873 |
Neg logit contributions | |
LLOW | -2.355 |
tten | -2.301 |
�� | -2.257 |
ADRA | -2.167 |
Swordsman | -2.165 |
Token | Good |
Token ID | 4599 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ippers | +0.973 |
ipper | +0.949 |
rones | +0.750 |
lems | +0.748 |
lihood | +0.737 |
Neg logit contributions | |
LLOW | -1.109 |
tten | -1.075 |
�� | -1.054 |
Swordsman | -1.028 |
FACE | -1.008 |
Token | Wife |
Token ID | 34188 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
LLOW | +1.215 |
tten | +1.177 |
�� | +1.148 |
Swordsman | +1.130 |
FACE | +1.102 |
Neg logit contributions | |
ippers | -1.248 |
ipper | -1.217 |
lems | -0.988 |
rones | -0.977 |
ploy | -0.968 |
Token | was |
Token ID | 373 |
Feature activation | -1.47e+00 |
Pos logit contributions | |
LLOW | +0.862 |
tten | +0.844 |
�� | +0.821 |
Swordsman | +0.796 |
FACE | +0.784 |
Neg logit contributions | |
ippers | -0.789 |
ipper | -0.773 |
lems | -0.615 |
rones | -0.609 |
Carney | -0.605 |
Token | complaint |
Token ID | 8224 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
ippers | +0.578 |
ipper | +0.564 |
rones | +0.454 |
lems | +0.453 |
Carney | +0.450 |
Neg logit contributions | |
LLOW | -0.569 |
tten | -0.553 |
�� | -0.542 |
Swordsman | -0.525 |
FACE | -0.520 |
Token | against |
Token ID | 1028 |
Feature activation | -2.36e+00 |
Pos logit contributions | |
ippers | +1.214 |
ipper | +1.183 |
rones | +0.934 |
lems | +0.930 |
Carney | +0.919 |
Neg logit contributions | |
LLOW | -1.396 |
tten | -1.364 |
�� | -1.336 |
Swordsman | -1.298 |
FACE | -1.281 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ippers | +1.675 |
ipper | +1.635 |
rones | +1.294 |
lems | +1.289 |
Carney | +1.276 |
Neg logit contributions | |
LLOW | -1.891 |
tten | -1.846 |
�� | -1.810 |
Swordsman | -1.752 |
ADRA | -1.733 |
Token | club |
Token ID | 3430 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
ippers | +0.826 |
ipper | +0.807 |
rones | +0.659 |
lems | +0.656 |
lihood | +0.649 |
Neg logit contributions | |
LLOW | -0.760 |
tten | -0.739 |
�� | -0.722 |
Swordsman | -0.702 |
FACE | -0.688 |
Token | on |
Token ID | 319 |
Feature activation | -9.22e-01 |
Pos logit contributions | |
ippers | +1.013 |
ipper | +0.990 |
rones | +0.792 |
lems | +0.789 |
Carney | +0.782 |
Neg logit contributions | |
LLOW | -1.055 |
tten | -1.029 |
�� | -1.007 |
Swordsman | -0.976 |
FACE | -0.963 |
Token | Oct |
Token ID | 2556 |
Feature activation | -6.77e+00 |
Pos logit contributions | |
ippers | +0.676 |
ipper | +0.659 |
lems | +0.527 |
rones | +0.525 |
lihood | +0.516 |
Neg logit contributions | |
LLOW | -0.719 |
tten | -0.702 |
�� | -0.686 |
Swordsman | -0.667 |
FACE | -0.660 |
Token | . |
Token ID | 13 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
ippers | +4.487 |
ipper | +4.428 |
rones | +3.510 |
lems | +3.445 |
ploy | +3.388 |
Neg logit contributions | |
LLOW | -5.531 |
�� | -5.451 |
tten | -5.427 |
ADRA | -5.263 |
Swordsman | -5.164 |
Token | 9 |
Token ID | 860 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
ippers | +1.551 |
ipper | +1.516 |
rones | +1.225 |
lems | +1.221 |
Carney | +1.208 |
Neg logit contributions | |
LLOW | -1.487 |
tten | -1.447 |
�� | -1.414 |
Swordsman | -1.370 |
FACE | -1.352 |
Token | , |
Token ID | 11 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
ippers | +1.466 |
ipper | +1.432 |
rones | +1.138 |
lems | +1.133 |
Carney | +1.122 |
Neg logit contributions | |
LLOW | -1.621 |
tten | -1.582 |
�� | -1.550 |
Swordsman | -1.499 |
ADRA | -1.485 |
Token | 2015 |
Token ID | 1853 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
ippers | +1.453 |
ipper | +1.416 |
rones | +1.105 |
lems | +1.101 |
Carney | +1.089 |
Neg logit contributions | |
LLOW | -1.793 |
tten | -1.752 |
�� | -1.719 |
Swordsman | -1.667 |
ADRA | -1.650 |
Token | . |
Token ID | 13 |
Feature activation | -8.18e-02 |
Pos logit contributions | |
ippers | +0.763 |
ipper | +0.746 |
rones | +0.599 |
lems | +0.597 |
Carney | +0.591 |
Neg logit contributions | |
LLOW | -0.771 |
tten | -0.751 |
�� | -0.736 |
Swordsman | -0.711 |
ADRA | -0.703 |
Token | education |
Token ID | 3707 |
Feature activation | -1.37e+00 |
Pos logit contributions | |
LLOW | +1.879 |
tten | +1.805 |
�� | +1.762 |
Swordsman | +1.725 |
ATH | +1.678 |
Neg logit contributions | |
ippers | -2.208 |
ipper | -2.169 |
lems | -1.772 |
mast | -1.760 |
rones | -1.756 |
Token | they |
Token ID | 484 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
ippers | +0.942 |
ipper | +0.918 |
rones | +0.723 |
lems | +0.719 |
Carney | +0.710 |
Neg logit contributions | |
LLOW | -1.127 |
tten | -1.103 |
�� | -1.081 |
Swordsman | -1.044 |
ADRA | -1.037 |
Token | can |
Token ID | 460 |
Feature activation | -6.40e-01 |
Pos logit contributions | |
ippers | +1.622 |
ipper | +1.584 |
rones | +1.260 |
lems | +1.255 |
Carney | +1.243 |
Neg logit contributions | |
LLOW | -1.763 |
tten | -1.720 |
�� | -1.685 |
Swordsman | -1.631 |
ADRA | -1.613 |
Token | and |
Token ID | 290 |
Feature activation | -3.08e-01 |
Pos logit contributions | |
ippers | +0.451 |
ipper | +0.440 |
rones | +0.347 |
lems | +0.346 |
Carney | +0.343 |
Neg logit contributions | |
LLOW | -0.515 |
tten | -0.503 |
�� | -0.493 |
Swordsman | -0.478 |
ADRA | -0.472 |
Token | then |
Token ID | 788 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
ippers | +0.165 |
ipper | +0.160 |
rones | +0.116 |
lems | +0.115 |
Carney | +0.113 |
Neg logit contributions | |
LLOW | -0.299 |
tten | -0.294 |
�� | -0.289 |
Swordsman | -0.282 |
ADRA | -0.279 |
Token | use |
Token ID | 779 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
LLOW | +0.431 |
tten | +0.419 |
�� | +0.410 |
Swordsman | +0.397 |
ADRA | +0.392 |
Neg logit contributions | |
ippers | -0.433 |
ipper | -0.423 |
rones | -0.341 |
lems | -0.340 |
Carney | -0.336 |
Token | it |
Token ID | 340 |
Feature activation | -1.39e-02 |
Pos logit contributions | |
LLOW | +3.423 |
tten | +3.292 |
�� | +3.149 |
Swordsman | +3.131 |
FACE | +3.115 |
Neg logit contributions | |
ippers | -3.724 |
ipper | -3.601 |
lems | -2.951 |
rones | -2.918 |
lihood | -2.881 |
Token | to |
Token ID | 284 |
Feature activation | -5.68e-01 |
Pos logit contributions | |
ippers | +0.011 |
ipper | +0.011 |
rones | +0.009 |
lems | +0.009 |
Carney | +0.009 |
Neg logit contributions | |
LLOW | -0.010 |
tten | -0.009 |
�� | -0.009 |
Swordsman | -0.009 |
FACE | -0.009 |
Token | ' |
Token ID | 705 |
Feature activation | -1.11e+00 |
Pos logit contributions | |
ippers | +0.455 |
ipper | +0.445 |
rones | +0.363 |
lems | +0.362 |
Carney | +0.358 |
Neg logit contributions | |
LLOW | -0.405 |
tten | -0.393 |
�� | -0.381 |
Swordsman | -0.370 |
FACE | -0.367 |
Token | lead |
Token ID | 28230 |
Feature activation | +1.23e+00 |
Pos logit contributions | |
ippers | +1.079 |
ipper | +1.059 |
rones | +0.899 |
lems | +0.897 |
Carney | +0.894 |
Neg logit contributions | |
LLOW | -0.597 |
tten | -0.575 |
�� | -0.553 |
Swordsman | -0.526 |
FACE | -0.524 |
Token | by |
Token ID | 416 |
Feature activation | +2.38e-01 |
Pos logit contributions | |
LLOW | +1.165 |
tten | +1.132 |
�� | +1.108 |
Swordsman | +1.092 |
FACE | +1.081 |
Neg logit contributions | |
ippers | -0.702 |
ipper | -0.679 |
lems | -0.496 |
rones | -0.492 |
mast | -0.484 |
Token | .[ |
Token ID | 3693 |
Feature activation | -4.23e-02 |
Pos logit contributions | |
ippers | +0.070 |
ipper | +0.069 |
rones | +0.055 |
lems | +0.055 |
Carney | +0.054 |
Neg logit contributions | |
LLOW | -0.070 |
tten | -0.068 |
�� | -0.067 |
Swordsman | -0.064 |
ADRA | -0.064 |
Token | 158 |
Token ID | 21273 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
ippers | +0.038 |
ipper | +0.037 |
rones | +0.031 |
lems | +0.031 |
Carney | +0.030 |
Neg logit contributions | |
LLOW | -0.026 |
tten | -0.026 |
�� | -0.025 |
Swordsman | -0.024 |
ADRA | -0.024 |
Token | ] |
Token ID | 60 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
LLOW | +0.915 |
tten | +0.888 |
�� | +0.866 |
Swordsman | +0.833 |
ADRA | +0.821 |
Neg logit contributions | |
ippers | -1.217 |
ipper | -1.193 |
rones | -0.990 |
lems | -0.986 |
Carney | -0.979 |
Token |
|
Token ID | 198 |
Feature activation | +2.32e+00 |
Pos logit contributions | |
LLOW | +0.241 |
tten | +0.235 |
�� | +0.229 |
Swordsman | +0.224 |
FACE | +0.221 |
Neg logit contributions | |
ippers | -0.224 |
ipper | -0.218 |
rones | -0.174 |
lems | -0.173 |
Carney | -0.170 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
LLOW | +1.753 |
tten | +1.687 |
FACE | +1.609 |
Swordsman | +1.599 |
�� | +1.594 |
Neg logit contributions | |
ippers | -1.743 |
ipper | -1.693 |
rones | -1.362 |
lems | -1.351 |
lihood | -1.348 |
Token | See |
Token ID | 6214 |
Feature activation | +4.85e+00 |
Pos logit contributions | |
LLOW | +1.107 |
tten | +1.073 |
�� | +1.027 |
FACE | +1.008 |
Swordsman | +0.981 |
Neg logit contributions | |
ippers | -1.615 |
ipper | -1.579 |
Carney | -1.322 |
lems | -1.317 |
rones | -1.317 |
Token | also |
Token ID | 635 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
LLOW | +2.944 |
tten | +2.843 |
�� | +2.720 |
Swordsman | +2.618 |
FACE | +2.596 |
Neg logit contributions | |
ippers | -4.323 |
ipper | -4.204 |
lems | -3.534 |
rones | -3.513 |
Carney | -3.476 |
Token | [ |
Token ID | 685 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
LLOW | +0.364 |
tten | +0.353 |
�� | +0.345 |
Swordsman | +0.329 |
ADRA | +0.327 |
Neg logit contributions | |
ippers | -0.542 |
ipper | -0.532 |
rones | -0.446 |
lems | -0.444 |
Carney | -0.442 |
Token | edit |
Token ID | 4370 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
LLOW | +1.164 |
tten | +1.127 |
�� | +1.067 |
Swordsman | +1.026 |
FACE | +1.024 |
Neg logit contributions | |
ippers | -2.152 |
ipper | -2.106 |
rones | -1.796 |
Carney | -1.793 |
lihood | -1.786 |
Token | ] |
Token ID | 2361 |
Feature activation | +3.87e+00 |
Pos logit contributions | |
LLOW | +0.646 |
tten | +0.623 |
�� | +0.601 |
Swordsman | +0.581 |
FACE | +0.568 |
Neg logit contributions | |
ippers | -1.102 |
ipper | -1.081 |
rones | -0.915 |
lems | -0.911 |
Carney | -0.904 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | +2.60e+00 |
Pos logit contributions | |
LLOW | +2.895 |
tten | +2.809 |
�� | +2.753 |
Swordsman | +2.659 |
FACE | +2.633 |
Neg logit contributions | |
ippers | -2.927 |
ipper | -2.860 |
rones | -2.305 |
lems | -2.279 |
ploy | -2.271 |
Token | northwest |
Token ID | 24821 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
LLOW | +1.092 |
tten | +1.070 |
�� | +1.035 |
Swordsman | +1.013 |
FACE | +1.001 |
Neg logit contributions | |
ippers | -1.041 |
ipper | -1.016 |
rones | -0.820 |
lems | -0.818 |
lihood | -0.803 |
Token | corner |
Token ID | 5228 |
Feature activation | +1.02e+00 |
Pos logit contributions | |
ippers | +0.991 |
ipper | +0.966 |
rones | +0.762 |
lems | +0.757 |
Carney | +0.754 |
Neg logit contributions | |
LLOW | -1.126 |
tten | -1.099 |
�� | -1.079 |
Swordsman | -1.043 |
ADRA | -1.037 |
Token | of |
Token ID | 286 |
Feature activation | +4.55e-02 |
Pos logit contributions | |
LLOW | +0.813 |
tten | +0.793 |
�� | +0.777 |
Swordsman | +0.752 |
FACE | +0.744 |
Neg logit contributions | |
ippers | -0.734 |
ipper | -0.716 |
rones | -0.569 |
lems | -0.567 |
Carney | -0.559 |
Token | St |
Token ID | 520 |
Feature activation | +2.94e-01 |
Pos logit contributions | |
LLOW | +0.035 |
tten | +0.034 |
�� | +0.033 |
Swordsman | +0.032 |
FACE | +0.032 |
Neg logit contributions | |
ippers | -0.034 |
ipper | -0.033 |
rones | -0.026 |
lems | -0.026 |
lihood | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +2.38e+00 |
Pos logit contributions | |
LLOW | +0.223 |
tten | +0.217 |
�� | +0.212 |
Swordsman | +0.205 |
ADRA | +0.203 |
Neg logit contributions | |
ippers | -0.222 |
ipper | -0.217 |
rones | -0.175 |
lems | -0.174 |
Carney | -0.173 |
Token | Clair |
Token ID | 48195 |
Feature activation | +4.54e+00 |
Pos logit contributions | |
LLOW | +1.607 |
tten | +1.559 |
�� | +1.523 |
Swordsman | +1.485 |
FACE | +1.450 |
Neg logit contributions | |
ippers | -1.965 |
ipper | -1.921 |
lems | -1.590 |
rones | -1.584 |
lihood | -1.568 |
Token | and |
Token ID | 290 |
Feature activation | +5.34e-01 |
Pos logit contributions | |
LLOW | +3.527 |
tten | +3.419 |
�� | +3.307 |
Swordsman | +3.304 |
FACE | +3.216 |
Neg logit contributions | |
ippers | -3.257 |
ipper | -3.180 |
rones | -2.554 |
lems | -2.515 |
lihood | -2.511 |
Token | Avenue |
Token ID | 8878 |
Feature activation | +2.29e+00 |
Pos logit contributions | |
LLOW | +0.413 |
tten | +0.403 |
�� | +0.392 |
Swordsman | +0.383 |
FACE | +0.377 |
Neg logit contributions | |
ippers | -0.395 |
ipper | -0.385 |
rones | -0.310 |
lems | -0.308 |
lihood | -0.303 |
Token | Road |
Token ID | 5567 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
LLOW | +1.779 |
tten | +1.739 |
�� | +1.687 |
Swordsman | +1.651 |
FACE | +1.634 |
Neg logit contributions | |
ippers | -1.676 |
ipper | -1.632 |
rones | -1.310 |
lems | -1.299 |
lihood | -1.286 |
Token |
|
Token ID | 198 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
LLOW | +1.857 |
tten | +1.812 |
�� | +1.762 |
Swordsman | +1.711 |
FACE | +1.692 |
Neg logit contributions | |
ippers | -1.665 |
ipper | -1.619 |
rones | -1.294 |
lems | -1.282 |
lihood | -1.271 |
Token |
|
Token ID | 198 |
Feature activation | +8.18e-01 |
Pos logit contributions | |
LLOW | +0.993 |
tten | +0.959 |
�� | +0.920 |
FACE | +0.911 |
Swordsman | +0.911 |
Neg logit contributions | |
ippers | -0.917 |
ipper | -0.893 |
rones | -0.711 |
lems | -0.707 |
lihood | -0.704 |
Token | rooting |
Token ID | 40105 |
Feature activation | +2.07e+00 |
Pos logit contributions | |
ippers | +0.145 |
ipper | +0.142 |
rones | +0.115 |
lems | +0.114 |
Carney | +0.112 |
Neg logit contributions | |
LLOW | -0.142 |
tten | -0.138 |
�� | -0.135 |
Swordsman | -0.131 |
FACE | -0.130 |
Token | for |
Token ID | 329 |
Feature activation | +3.04e-01 |
Pos logit contributions | |
LLOW | +1.810 |
tten | +1.752 |
�� | +1.711 |
Swordsman | +1.670 |
FACE | +1.658 |
Neg logit contributions | |
ippers | -1.308 |
ipper | -1.288 |
rones | -0.995 |
lihood | -0.968 |
Carney | -0.966 |
Token | the |
Token ID | 262 |
Feature activation | +3.25e-01 |
Pos logit contributions | |
LLOW | +0.220 |
tten | +0.214 |
�� | +0.209 |
Swordsman | +0.203 |
FACE | +0.199 |
Neg logit contributions | |
ippers | -0.239 |
ipper | -0.234 |
rones | -0.190 |
lems | -0.189 |
lihood | -0.187 |
Token | Democratic |
Token ID | 4390 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
LLOW | +0.258 |
tten | +0.252 |
�� | +0.246 |
Swordsman | +0.239 |
FACE | +0.236 |
Neg logit contributions | |
ippers | -0.232 |
ipper | -0.227 |
rones | -0.181 |
lems | -0.180 |
lihood | -0.177 |
Token | presidential |
Token ID | 4787 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
LLOW | +1.389 |
tten | +1.363 |
�� | +1.323 |
Swordsman | +1.301 |
FACE | +1.270 |
Neg logit contributions | |
ippers | -1.293 |
ipper | -1.269 |
rones | -1.019 |
lems | -1.013 |
Carney | -1.004 |
Token | candidate |
Token ID | 4540 |
Feature activation | +4.90e+00 |
Pos logit contributions | |
LLOW | +1.208 |
tten | +1.190 |
�� | +1.158 |
Swordsman | +1.136 |
ADRA | +1.115 |
Neg logit contributions | |
ippers | -0.954 |
ipper | -0.932 |
rones | -0.730 |
lems | -0.727 |
Carney | -0.721 |
Token | say |
Token ID | 910 |
Feature activation | +5.99e-01 |
Pos logit contributions | |
LLOW | +3.542 |
tten | +3.467 |
Swordsman | +3.300 |
FACE | +3.296 |
�� | +3.261 |
Neg logit contributions | |
ippers | -3.705 |
ipper | -3.649 |
rones | -2.972 |
lems | -2.938 |
ploy | -2.876 |
Token | their |
Token ID | 511 |
Feature activation | +1.04e+00 |
Pos logit contributions | |
LLOW | +0.466 |
tten | +0.454 |
�� | +0.441 |
Swordsman | +0.432 |
FACE | +0.426 |
Neg logit contributions | |
ippers | -0.438 |
ipper | -0.428 |
rones | -0.344 |
lems | -0.340 |
lihood | -0.337 |
Token | accounts |
Token ID | 5504 |
Feature activation | +9.47e-01 |
Pos logit contributions | |
LLOW | +0.804 |
tten | +0.786 |
�� | +0.754 |
Swordsman | +0.740 |
FACE | +0.738 |
Neg logit contributions | |
ippers | -0.766 |
ipper | -0.754 |
rones | -0.601 |
lems | -0.594 |
ploy | -0.591 |
Token | on |
Token ID | 319 |
Feature activation | -1.20e-02 |
Pos logit contributions | |
LLOW | +0.733 |
tten | +0.718 |
�� | +0.699 |
Swordsman | +0.679 |
FACE | +0.676 |
Neg logit contributions | |
ippers | -0.692 |
ipper | -0.675 |
lems | -0.543 |
rones | -0.543 |
lihood | -0.531 |
Token | the |
Token ID | 262 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
ippers | +0.010 |
ipper | +0.009 |
rones | +0.008 |
lems | +0.008 |
lihood | +0.007 |
Neg logit contributions | |
LLOW | -0.009 |
tten | -0.008 |
�� | -0.008 |
Swordsman | -0.008 |
FACE | -0.008 |
Token | handle |
Token ID | 5412 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
LLOW | +0.624 |
tten | +0.618 |
�� | +0.590 |
Swordsman | +0.579 |
FACE | +0.572 |
Neg logit contributions | |
ippers | -0.571 |
ipper | -0.553 |
rones | -0.445 |
lems | -0.443 |
lihood | -0.436 |
Token | multiple |
Token ID | 3294 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
ippers | +1.356 |
ipper | +1.326 |
rones | +1.060 |
lems | +1.056 |
Carney | +1.048 |
Neg logit contributions | |
LLOW | -1.417 |
tten | -1.382 |
�� | -1.355 |
Swordsman | -1.309 |
ADRA | -1.295 |
Token | spikes |
Token ID | 27198 |
Feature activation | +2.20e+00 |
Pos logit contributions | |
ippers | +1.340 |
ipper | +1.310 |
rones | +1.051 |
lems | +1.047 |
Carney | +1.037 |
Neg logit contributions | |
LLOW | -1.370 |
tten | -1.336 |
�� | -1.309 |
Swordsman | -1.265 |
ADRA | -1.251 |
Token | in |
Token ID | 287 |
Feature activation | +1.00e-02 |
Pos logit contributions | |
LLOW | +1.666 |
tten | +1.626 |
�� | +1.568 |
Swordsman | +1.546 |
FACE | +1.510 |
Neg logit contributions | |
ippers | -1.636 |
ipper | -1.598 |
rones | -1.297 |
lems | -1.293 |
mast | -1.264 |
Token | the |
Token ID | 262 |
Feature activation | +6.76e-01 |
Pos logit contributions | |
LLOW | +0.008 |
tten | +0.008 |
�� | +0.007 |
Swordsman | +0.007 |
FACE | +0.007 |
Neg logit contributions | |
ippers | -0.007 |
ipper | -0.007 |
lems | -0.006 |
rones | -0.006 |
Carney | -0.006 |
Token | angle |
Token ID | 9848 |
Feature activation | +5.50e+00 |
Pos logit contributions | |
LLOW | +0.521 |
tten | +0.504 |
�� | +0.492 |
Swordsman | +0.479 |
FACE | +0.476 |
Neg logit contributions | |
ippers | -0.504 |
ipper | -0.487 |
lems | -0.394 |
rones | -0.390 |
ploy | -0.385 |
Token | - |
Token ID | 12 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
LLOW | +4.115 |
tten | +3.978 |
�� | +3.900 |
Swordsman | +3.783 |
FACE | +3.758 |
Neg logit contributions | |
ippers | -4.093 |
ipper | -3.941 |
lems | -3.240 |
rones | -3.185 |
lihood | -3.159 |
Token | of |
Token ID | 1659 |
Feature activation | -5.12e-02 |
Pos logit contributions | |
LLOW | +1.008 |
tten | +0.994 |
�� | +0.951 |
FACE | +0.928 |
Swordsman | +0.923 |
Neg logit contributions | |
ippers | -1.002 |
ipper | -0.974 |
rones | -0.788 |
Carney | -0.788 |
lems | -0.783 |
Token | - |
Token ID | 12 |
Feature activation | -6.87e-02 |
Pos logit contributions | |
ippers | +0.034 |
ipper | +0.033 |
rones | +0.026 |
lems | +0.026 |
Carney | +0.026 |
Neg logit contributions | |
LLOW | -0.043 |
tten | -0.042 |
�� | -0.041 |
Swordsman | -0.040 |
FACE | -0.040 |
Token | attack |
Token ID | 20358 |
Feature activation | -5.13e-01 |
Pos logit contributions | |
ippers | +0.049 |
ipper | +0.048 |
rones | +0.038 |
lems | +0.038 |
Carney | +0.038 |
Neg logit contributions | |
LLOW | -0.054 |
tten | -0.053 |
�� | -0.052 |
Swordsman | -0.050 |
FACE | -0.050 |
Token | data |
Token ID | 1366 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
ippers | +0.401 |
ipper | +0.392 |
rones | +0.318 |
lems | +0.318 |
Carney | +0.313 |
Neg logit contributions | |
LLOW | -0.374 |
tten | -0.364 |
�� | -0.356 |
Swordsman | -0.344 |
FACE | -0.340 |
Token | be |
Token ID | 1350 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
ippers | +0.326 |
ipper | +0.319 |
rones | +0.258 |
lems | +0.257 |
Carney | +0.255 |
Neg logit contributions | |
LLOW | -0.321 |
tten | -0.315 |
�� | -0.305 |
Swordsman | -0.296 |
FACE | -0.293 |
Token | 09 |
Token ID | 2931 |
Feature activation | +1.25e+00 |
Pos logit contributions | |
ippers | +1.925 |
ipper | +1.884 |
rones | +1.526 |
lems | +1.523 |
Carney | +1.504 |
Neg logit contributions | |
LLOW | -1.791 |
tten | -1.740 |
�� | -1.713 |
Swordsman | -1.647 |
ADRA | -1.628 |
Token | c |
Token ID | 66 |
Feature activation | -9.55e-01 |
Pos logit contributions | |
LLOW | +0.902 |
tten | +0.889 |
�� | +0.849 |
Swordsman | +0.830 |
FACE | +0.822 |
Neg logit contributions | |
ippers | -0.974 |
ipper | -0.950 |
rones | -0.778 |
Carney | -0.775 |
lems | -0.771 |
Token | 68 |
Token ID | 3104 |
Feature activation | +9.13e-01 |
Pos logit contributions | |
ippers | +0.765 |
ipper | +0.749 |
rones | +0.612 |
lems | +0.610 |
Carney | +0.605 |
Neg logit contributions | |
LLOW | -0.676 |
tten | -0.658 |
�� | -0.642 |
Swordsman | -0.620 |
ADRA | -0.612 |
Token | a |
Token ID | 64 |
Feature activation | +6.42e-01 |
Pos logit contributions | |
LLOW | +0.660 |
tten | +0.651 |
�� | +0.624 |
Swordsman | +0.609 |
FACE | +0.601 |
Neg logit contributions | |
ippers | -0.709 |
ipper | -0.693 |
rones | -0.566 |
Carney | -0.563 |
lems | -0.562 |
Token | 410 |
Token ID | 33289 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
LLOW | +0.448 |
tten | +0.437 |
�� | +0.422 |
Swordsman | +0.410 |
FACE | +0.406 |
Neg logit contributions | |
ippers | -0.518 |
ipper | -0.506 |
rones | -0.416 |
lems | -0.414 |
Carney | -0.414 |
Token | f |
Token ID | 69 |
Feature activation | -2.11e-01 |
Pos logit contributions | |
LLOW | +1.613 |
tten | +1.600 |
�� | +1.514 |
Swordsman | +1.491 |
FACE | +1.478 |
Neg logit contributions | |
ippers | -1.731 |
ipper | -1.686 |
rones | -1.390 |
Carney | -1.382 |
lems | -1.376 |
Token | 6 |
Token ID | 21 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
ippers | +0.170 |
ipper | +0.167 |
rones | +0.137 |
lems | +0.136 |
Carney | +0.136 |
Neg logit contributions | |
LLOW | -0.147 |
tten | -0.144 |
�� | -0.139 |
Swordsman | -0.135 |
ADRA | -0.133 |
Token | bed |
Token ID | 3077 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
LLOW | +1.372 |
tten | +1.361 |
�� | +1.303 |
Swordsman | +1.276 |
FACE | +1.257 |
Neg logit contributions | |
ippers | -1.319 |
ipper | -1.285 |
rones | -1.045 |
Carney | -1.031 |
lems | -1.031 |
Token | 15 |
Token ID | 1314 |
Feature activation | +2.30e+00 |
Pos logit contributions | |
LLOW | +1.134 |
tten | +1.119 |
�� | +1.066 |
Swordsman | +1.043 |
FACE | +1.036 |
Neg logit contributions | |
ippers | -1.220 |
ipper | -1.190 |
rones | -0.974 |
Carney | -0.972 |
lems | -0.965 |
Token | c |
Token ID | 66 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
LLOW | +1.641 |
tten | +1.630 |
�� | +1.537 |
Swordsman | +1.509 |
FACE | +1.497 |
Neg logit contributions | |
ippers | -1.777 |
ipper | -1.736 |
Carney | -1.428 |
rones | -1.424 |
lems | -1.414 |
Token | . |
Token ID | 13 |
Feature activation | +2.78e-01 |
Pos logit contributions | |
LLOW | +1.306 |
tten | +1.284 |
�� | +1.244 |
Swordsman | +1.227 |
FACE | +1.201 |
Neg logit contributions | |
ippers | -1.148 |
ipper | -1.115 |
lems | -0.880 |
rones | -0.878 |
Carney | -0.872 |
Token |
|
Token ID | 198 |
Feature activation | +7.56e-01 |
Pos logit contributions | |
LLOW | +0.215 |
tten | +0.209 |
�� | +0.203 |
Swordsman | +0.202 |
FACE | +0.197 |
Neg logit contributions | |
ippers | -0.205 |
ipper | -0.201 |
rones | -0.159 |
lems | -0.159 |
ploy | -0.157 |
Token |
|
Token ID | 198 |
Feature activation | +3.70e-01 |
Pos logit contributions | |
LLOW | +0.603 |
tten | +0.585 |
�� | +0.563 |
Swordsman | +0.555 |
FACE | +0.553 |
Neg logit contributions | |
ippers | -0.540 |
ipper | -0.525 |
rones | -0.417 |
lems | -0.415 |
Carney | -0.412 |
Token | However |
Token ID | 4864 |
Feature activation | -4.29e-01 |
Pos logit contributions | |
LLOW | +0.246 |
tten | +0.238 |
�� | +0.228 |
FACE | +0.225 |
ATH | +0.220 |
Neg logit contributions | |
ippers | -0.314 |
ipper | -0.307 |
Carney | -0.254 |
rones | -0.253 |
lems | -0.252 |
Token | , |
Token ID | 11 |
Feature activation | +7.89e-01 |
Pos logit contributions | |
ippers | +0.306 |
ipper | +0.299 |
rones | +0.237 |
lems | +0.236 |
Carney | +0.233 |
Neg logit contributions | |
LLOW | -0.341 |
tten | -0.335 |
�� | -0.328 |
Swordsman | -0.317 |
ADRA | -0.312 |
Token | now |
Token ID | 783 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
LLOW | +0.609 |
tten | +0.603 |
�� | +0.583 |
Swordsman | +0.574 |
FACE | +0.554 |
Neg logit contributions | |
ippers | -0.573 |
ipper | -0.560 |
rones | -0.447 |
lems | -0.446 |
ploy | -0.445 |
Token | the |
Token ID | 262 |
Feature activation | +1.48e+00 |
Pos logit contributions | |
LLOW | +1.142 |
tten | +1.130 |
�� | +1.087 |
Swordsman | +1.063 |
FACE | +1.038 |
Neg logit contributions | |
ippers | -1.087 |
ipper | -1.066 |
lems | -0.860 |
rones | -0.853 |
mast | -0.840 |
Token | tide |
Token ID | 20013 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
LLOW | +1.144 |
tten | +1.125 |
�� | +1.095 |
Swordsman | +1.069 |
FACE | +1.041 |
Neg logit contributions | |
ippers | -1.068 |
ipper | -1.042 |
lems | -0.829 |
ploy | -0.828 |
rones | -0.826 |
Token | has |
Token ID | 468 |
Feature activation | +6.25e-01 |
Pos logit contributions | |
LLOW | +2.862 |
tten | +2.804 |
�� | +2.753 |
Swordsman | +2.677 |
FACE | +2.614 |
Neg logit contributions | |
ippers | -2.535 |
ipper | -2.468 |
rones | -1.950 |
lems | -1.937 |
lihood | -1.932 |
Token | turned |
Token ID | 2900 |
Feature activation | +1.08e+00 |
Pos logit contributions | |
LLOW | +0.514 |
tten | +0.504 |
�� | +0.495 |
Swordsman | +0.477 |
FACE | +0.472 |
Neg logit contributions | |
ippers | -0.426 |
ipper | -0.415 |
rones | -0.328 |
lems | -0.323 |
lihood | -0.322 |
Token | again |
Token ID | 757 |
Feature activation | +9.96e-01 |
Pos logit contributions | |
LLOW | +0.813 |
tten | +0.794 |
�� | +0.776 |
Swordsman | +0.754 |
FACE | +0.740 |
Neg logit contributions | |
ippers | -0.809 |
ipper | -0.789 |
rones | -0.636 |
lems | -0.627 |
ploy | -0.626 |
Token | ? |
Token ID | 30 |
Feature activation | +2.99e-01 |
Pos logit contributions | |
LLOW | +1.102 |
tten | +1.083 |
�� | +1.039 |
Swordsman | +1.027 |
ADRA | +1.009 |
Neg logit contributions | |
ippers | -1.164 |
ipper | -1.140 |
rones | -0.923 |
Carney | -0.914 |
lems | -0.907 |
Token | What |
Token ID | 1867 |
Feature activation | +1.21e+00 |
Pos logit contributions | |
LLOW | +0.241 |
tten | +0.234 |
�� | +0.229 |
Swordsman | +0.224 |
FACE | +0.221 |
Neg logit contributions | |
ippers | -0.213 |
ipper | -0.207 |
rones | -0.164 |
lems | -0.163 |
Carney | -0.161 |
Token | is |
Token ID | 318 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
LLOW | +0.901 |
tten | +0.882 |
�� | +0.850 |
Swordsman | +0.836 |
ADRA | +0.825 |
Neg logit contributions | |
ippers | -0.917 |
ipper | -0.897 |
rones | -0.726 |
Carney | -0.717 |
lems | -0.714 |
Token | it |
Token ID | 340 |
Feature activation | +9.82e-01 |
Pos logit contributions | |
LLOW | +1.603 |
tten | +1.560 |
�� | +1.515 |
Swordsman | +1.493 |
FACE | +1.467 |
Neg logit contributions | |
ippers | -1.607 |
ipper | -1.567 |
lems | -1.272 |
rones | -1.268 |
lihood | -1.245 |
Token | ? |
Token ID | 30 |
Feature activation | -4.45e-01 |
Pos logit contributions | |
LLOW | +0.682 |
tten | +0.668 |
�� | +0.646 |
Swordsman | +0.632 |
FACE | +0.622 |
Neg logit contributions | |
ippers | -0.802 |
ipper | -0.781 |
rones | -0.641 |
lems | -0.638 |
Carney | -0.632 |
Token | I |
Token ID | 314 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
ippers | +0.309 |
ipper | +0.301 |
rones | +0.237 |
lems | +0.236 |
Carney | +0.233 |
Neg logit contributions | |
LLOW | -0.364 |
tten | -0.355 |
�� | -0.348 |
Swordsman | -0.338 |
FACE | -0.335 |
Token | � |
Token ID | 447 |
Feature activation | -3.63e-01 |
Pos logit contributions | |
LLOW | +0.451 |
tten | +0.440 |
�� | +0.426 |
Swordsman | +0.415 |
FACE | +0.409 |
Neg logit contributions | |
ippers | -0.458 |
ipper | -0.447 |
rones | -0.362 |
lems | -0.360 |
Carney | -0.357 |
Token | � |
Token ID | 247 |
Feature activation | +5.38e-01 |
Pos logit contributions | |
ippers | +0.262 |
ipper | +0.256 |
rones | +0.206 |
lems | +0.205 |
lihood | +0.203 |
Neg logit contributions | |
LLOW | -0.285 |
tten | -0.278 |
�� | -0.276 |
Swordsman | -0.262 |
FACE | -0.260 |
Token | ll |
Token ID | 297 |
Feature activation | +1.46e+00 |
Pos logit contributions | |
LLOW | +0.429 |
tten | +0.424 |
�� | +0.409 |
Swordsman | +0.397 |
FACE | +0.394 |
Neg logit contributions | |
ippers | -0.383 |
ipper | -0.372 |
rones | -0.297 |
lems | -0.293 |
Carney | -0.292 |
Token | definitely |
Token ID | 4753 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
LLOW | +1.149 |
tten | +1.128 |
�� | +1.090 |
Swordsman | +1.063 |
FACE | +1.055 |
Neg logit contributions | |
ippers | -1.062 |
ipper | -1.031 |
rones | -0.828 |
lems | -0.815 |
lihood | -0.814 |
Token | grant |
Token ID | 7264 |
Feature activation | -4.15e-01 |
Pos logit contributions | |
LLOW | +0.716 |
tten | +0.704 |
�� | +0.676 |
Swordsman | +0.663 |
FACE | +0.655 |
Neg logit contributions | |
ippers | -0.676 |
ipper | -0.655 |
rones | -0.528 |
lems | -0.522 |
lihood | -0.518 |
Token | arrive |
Token ID | 9240 |
Feature activation | +1.37e+00 |
Pos logit contributions | |
ippers | +0.326 |
ipper | +0.317 |
rones | +0.254 |
lems | +0.252 |
Carney | +0.249 |
Neg logit contributions | |
LLOW | -0.354 |
tten | -0.348 |
�� | -0.337 |
Swordsman | -0.329 |
FACE | -0.323 |
Token | in |
Token ID | 287 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
LLOW | +1.024 |
tten | +1.011 |
�� | +0.968 |
Swordsman | +0.957 |
FACE | +0.931 |
Neg logit contributions | |
ippers | -1.038 |
ipper | -1.007 |
rones | -0.815 |
lihood | -0.807 |
lems | -0.806 |
Token | an |
Token ID | 281 |
Feature activation | +1.94e+00 |
Pos logit contributions | |
LLOW | +0.913 |
tten | +0.897 |
�� | +0.868 |
Swordsman | +0.848 |
ADRA | +0.821 |
Neg logit contributions | |
ippers | -0.960 |
ipper | -0.934 |
rones | -0.762 |
lems | -0.756 |
lihood | -0.753 |
Token | hour |
Token ID | 1711 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
LLOW | +1.193 |
tten | +1.168 |
�� | +1.149 |
Swordsman | +1.105 |
FACE | +1.063 |
Neg logit contributions | |
ippers | -1.743 |
ipper | -1.693 |
rones | -1.426 |
lems | -1.397 |
lihood | -1.394 |
Token | or |
Token ID | 393 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
LLOW | +1.408 |
tten | +1.368 |
�� | +1.331 |
Swordsman | +1.286 |
FACE | +1.260 |
Neg logit contributions | |
ippers | -1.749 |
ipper | -1.705 |
rones | -1.403 |
lems | -1.392 |
Carney | -1.391 |
Token | two |
Token ID | 734 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
LLOW | +1.302 |
tten | +1.274 |
�� | +1.234 |
Swordsman | +1.211 |
FACE | +1.183 |
Neg logit contributions | |
ippers | -1.415 |
ipper | -1.376 |
rones | -1.125 |
lems | -1.113 |
Carney | -1.099 |
Token | . |
Token ID | 13 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
LLOW | +2.575 |
tten | +2.489 |
�� | +2.415 |
Swordsman | +2.331 |
FACE | +2.287 |
Neg logit contributions | |
ippers | -3.376 |
ipper | -3.285 |
rones | -2.751 |
lihood | -2.692 |
Carney | -2.686 |
Token | The |
Token ID | 383 |
Feature activation | +1.09e+00 |
Pos logit contributions | |
LLOW | +1.100 |
tten | +1.050 |
�� | +1.029 |
Swordsman | +1.024 |
FACE | +1.002 |
Neg logit contributions | |
ippers | -1.073 |
ipper | -1.046 |
rones | -0.845 |
lems | -0.836 |
ploy | -0.822 |
Token | outside |
Token ID | 2354 |
Feature activation | +4.69e-01 |
Pos logit contributions | |
LLOW | +0.816 |
tten | +0.803 |
�� | +0.776 |
Swordsman | +0.767 |
FACE | +0.739 |
Neg logit contributions | |
ippers | -0.825 |
ipper | -0.807 |
rones | -0.651 |
lihood | -0.644 |
lems | -0.642 |
Token | is |
Token ID | 318 |
Feature activation | +5.80e-01 |
Pos logit contributions | |
LLOW | +0.360 |
tten | +0.353 |
�� | +0.343 |
Swordsman | +0.337 |
FACE | +0.331 |
Neg logit contributions | |
ippers | -0.345 |
ipper | -0.340 |
rones | -0.273 |
lems | -0.270 |
ploy | -0.267 |
Token | becoming |
Token ID | 5033 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
LLOW | +0.438 |
tten | +0.435 |
�� | +0.422 |
Swordsman | +0.407 |
FACE | +0.403 |
Neg logit contributions | |
ippers | -0.436 |
ipper | -0.426 |
rones | -0.344 |
lems | -0.339 |
lihood | -0.335 |
Token |
|
Token ID | 198 |
Feature activation | -1.32e-01 |
Pos logit contributions | |
LLOW | +0.176 |
tten | +0.171 |
�� | +0.165 |
Swordsman | +0.162 |
FACE | +0.161 |
Neg logit contributions | |
ippers | -0.153 |
ipper | -0.149 |
rones | -0.118 |
lems | -0.117 |
lihood | -0.116 |
Token | " |
Token ID | 1 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
ippers | +0.112 |
ipper | +0.109 |
rones | +0.090 |
lems | +0.090 |
Carney | +0.089 |
Neg logit contributions | |
LLOW | -0.088 |
tten | -0.085 |
�� | -0.083 |
FACE | -0.080 |
ADRA | -0.079 |
Token | For |
Token ID | 1890 |
Feature activation | +3.07e+00 |
Pos logit contributions | |
ippers | +1.133 |
ipper | +1.110 |
rones | +0.911 |
lems | +0.909 |
Carney | +0.900 |
Neg logit contributions | |
LLOW | -0.931 |
tten | -0.901 |
�� | -0.881 |
Swordsman | -0.847 |
FACE | -0.840 |
Token | all |
Token ID | 477 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
LLOW | +2.312 |
tten | +2.286 |
�� | +2.222 |
ADRA | +2.139 |
Swordsman | +2.127 |
Neg logit contributions | |
ippers | -2.274 |
ipper | -2.225 |
rones | -1.777 |
lems | -1.768 |
mast | -1.748 |
Token | the |
Token ID | 262 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
LLOW | +1.495 |
tten | +1.463 |
�� | +1.428 |
Swordsman | +1.383 |
ADRA | +1.375 |
Neg logit contributions | |
ippers | -1.379 |
ipper | -1.348 |
rones | -1.076 |
lems | -1.075 |
lihood | -1.050 |
Token | pomp |
Token ID | 49435 |
Feature activation | +2.65e+00 |
Pos logit contributions | |
LLOW | +0.724 |
tten | +0.710 |
�� | +0.693 |
Swordsman | +0.670 |
FACE | +0.665 |
Neg logit contributions | |
ippers | -0.746 |
ipper | -0.730 |
rones | -0.590 |
lems | -0.588 |
lihood | -0.574 |
Token | and |
Token ID | 290 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
LLOW | +2.052 |
tten | +1.997 |
�� | +1.957 |
Swordsman | +1.888 |
FACE | +1.884 |
Neg logit contributions | |
ippers | -1.920 |
ipper | -1.877 |
lems | -1.496 |
rones | -1.474 |
lihood | -1.465 |
Token | ceremony |
Token ID | 10542 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
ippers | +0.242 |
ipper | +0.237 |
rones | +0.190 |
lems | +0.189 |
Carney | +0.187 |
Neg logit contributions | |
LLOW | -0.248 |
tten | -0.242 |
�� | -0.236 |
Swordsman | -0.229 |
FACE | -0.227 |
Token | , |
Token ID | 11 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
LLOW | +0.797 |
tten | +0.778 |
�� | +0.761 |
Swordsman | +0.737 |
FACE | +0.732 |
Neg logit contributions | |
ippers | -0.749 |
ipper | -0.732 |
lems | -0.587 |
rones | -0.584 |
lihood | -0.571 |
Token | nothing |
Token ID | 2147 |
Feature activation | +2.42e+00 |
Pos logit contributions | |
LLOW | +1.297 |
tten | +1.265 |
�� | +1.234 |
Swordsman | +1.205 |
FACE | +1.192 |
Neg logit contributions | |
ippers | -1.082 |
ipper | -1.054 |
lems | -0.832 |
rones | -0.823 |
lihood | -0.804 |
Token | in |
Token ID | 287 |
Feature activation | +7.03e-01 |
Pos logit contributions | |
LLOW | +1.847 |
tten | +1.801 |
�� | +1.769 |
Swordsman | +1.725 |
FACE | +1.701 |
Neg logit contributions | |
ippers | -1.810 |
ipper | -1.767 |
lems | -1.409 |
rones | -1.395 |
mast | -1.374 |
Token | car |
Token ID | 7718 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
LLOW | +1.272 |
tten | +1.265 |
�� | +1.217 |
FACE | +1.199 |
Swordsman | +1.190 |
Neg logit contributions | |
ippers | -0.735 |
ipper | -0.713 |
lihood | -0.516 |
rones | -0.510 |
lems | -0.506 |
Token | ibe |
Token ID | 32438 |
Feature activation | -2.22e+00 |
Pos logit contributions | |
ippers | +1.100 |
ipper | +1.042 |
rones | +0.539 |
lems | +0.535 |
Carney | +0.513 |
Neg logit contributions | |
LLOW | -4.127 |
tten | -4.052 |
�� | -4.009 |
Swordsman | -3.912 |
ADRA | -3.899 |
Token | deal |
Token ID | 1730 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
ippers | +1.246 |
ipper | +1.207 |
rones | +0.888 |
lems | +0.884 |
Carney | +0.865 |
Neg logit contributions | |
LLOW | -2.105 |
tten | -2.068 |
�� | -2.027 |
Swordsman | -1.977 |
ADRA | -1.957 |
Token | would |
Token ID | 561 |
Feature activation | -2.14e+00 |
Pos logit contributions | |
ippers | +0.875 |
ipper | +0.854 |
lems | +0.672 |
rones | +0.669 |
ploy | +0.655 |
Neg logit contributions | |
LLOW | -1.009 |
tten | -0.992 |
�� | -0.969 |
Swordsman | -0.938 |
FACE | -0.927 |
Token | help |
Token ID | 1037 |
Feature activation | +2.10e-01 |
Pos logit contributions | |
ippers | +1.536 |
ipper | +1.499 |
rones | +1.194 |
lems | +1.189 |
Carney | +1.167 |
Neg logit contributions | |
LLOW | -1.697 |
tten | -1.666 |
�� | -1.618 |
Swordsman | -1.572 |
FACE | -1.557 |
Token | lower |
Token ID | 2793 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
LLOW | +0.163 |
tten | +0.160 |
�� | +0.157 |
Swordsman | +0.151 |
ADRA | +0.150 |
Neg logit contributions | |
ippers | -0.153 |
ipper | -0.150 |
rones | -0.120 |
lems | -0.119 |
mast | -0.118 |
Token | and |
Token ID | 290 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
ippers | +1.796 |
ipper | +1.752 |
rones | +1.391 |
lems | +1.385 |
Carney | +1.373 |
Neg logit contributions | |
LLOW | -1.983 |
tten | -1.935 |
�� | -1.897 |
Swordsman | -1.836 |
ADRA | -1.816 |
Token | stabilize |
Token ID | 32411 |
Feature activation | -1.12e+00 |
Pos logit contributions | |
ippers | +0.880 |
ipper | +0.862 |
lems | +0.686 |
rones | +0.681 |
Carney | +0.674 |
Neg logit contributions | |
LLOW | -0.942 |
tten | -0.923 |
�� | -0.905 |
Swordsman | -0.871 |
FACE | -0.859 |
Token | the |
Token ID | 262 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
ippers | +0.818 |
ipper | +0.800 |
rones | +0.639 |
lems | +0.637 |
Carney | +0.625 |
Neg logit contributions | |
LLOW | -0.878 |
tten | -0.858 |
�� | -0.841 |
Swordsman | -0.811 |
FACE | -0.801 |
Token | cost |
Token ID | 1575 |
Feature activation | +2.19e+00 |
Pos logit contributions | |
ippers | +0.750 |
ipper | +0.732 |
rones | +0.581 |
lems | +0.580 |
mast | +0.569 |
Neg logit contributions | |
LLOW | -0.838 |
tten | -0.818 |
�� | -0.804 |
Swordsman | -0.776 |
FACE | -0.765 |
Token | of |
Token ID | 286 |
Feature activation | -3.13e-01 |
Pos logit contributions | |
LLOW | +1.628 |
tten | +1.586 |
�� | +1.549 |
Swordsman | +1.495 |
FACE | +1.493 |
Neg logit contributions | |
ippers | -1.654 |
ipper | -1.618 |
rones | -1.308 |
lems | -1.307 |
mast | -1.288 |
Token | by |
Token ID | 416 |
Feature activation | -1.31e+00 |
Pos logit contributions | |
ippers | +0.093 |
ipper | +0.090 |
rones | +0.069 |
lems | +0.068 |
Carney | +0.067 |
Neg logit contributions | |
LLOW | -0.137 |
tten | -0.133 |
�� | -0.131 |
Swordsman | -0.128 |
FACE | -0.126 |
Token | trying |
Token ID | 2111 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
ippers | +0.999 |
ipper | +0.977 |
rones | +0.789 |
lems | +0.784 |
Carney | +0.775 |
Neg logit contributions | |
LLOW | -0.987 |
tten | -0.961 |
�� | -0.939 |
Swordsman | -0.909 |
FACE | -0.897 |
Token | to |
Token ID | 284 |
Feature activation | -2.10e-01 |
Pos logit contributions | |
LLOW | +1.997 |
tten | +1.943 |
�� | +1.900 |
Swordsman | +1.833 |
FACE | +1.815 |
Neg logit contributions | |
ippers | -2.355 |
ipper | -2.308 |
rones | -1.892 |
Carney | -1.869 |
lems | -1.869 |
Token | reconcile |
Token ID | 33570 |
Feature activation | -2.32e+00 |
Pos logit contributions | |
ippers | +0.155 |
ipper | +0.152 |
rones | +0.123 |
lems | +0.121 |
Carney | +0.120 |
Neg logit contributions | |
LLOW | -0.163 |
tten | -0.159 |
�� | -0.153 |
Swordsman | -0.150 |
FACE | -0.149 |
Token | that |
Token ID | 326 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
ippers | +1.631 |
ipper | +1.594 |
rones | +1.259 |
lems | +1.252 |
Carney | +1.243 |
Neg logit contributions | |
LLOW | -1.864 |
tten | -1.820 |
�� | -1.788 |
Swordsman | -1.725 |
ADRA | -1.713 |
Token | with |
Token ID | 351 |
Feature activation | -4.30e-01 |
Pos logit contributions | |
ippers | +0.198 |
ipper | +0.193 |
rones | +0.154 |
lems | +0.153 |
Carney | +0.151 |
Neg logit contributions | |
LLOW | -0.216 |
tten | -0.211 |
�� | -0.205 |
Swordsman | -0.200 |
FACE | -0.198 |
Token | his |
Token ID | 465 |
Feature activation | +9.34e-01 |
Pos logit contributions | |
ippers | +0.326 |
ipper | +0.319 |
rones | +0.257 |
lems | +0.256 |
Carney | +0.253 |
Neg logit contributions | |
LLOW | -0.323 |
tten | -0.315 |
�� | -0.308 |
Swordsman | -0.298 |
FACE | -0.294 |
Token | religion |
Token ID | 5737 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
LLOW | +0.681 |
tten | +0.665 |
�� | +0.642 |
Swordsman | +0.626 |
FACE | +0.616 |
Neg logit contributions | |
ippers | -0.731 |
ipper | -0.714 |
rones | -0.582 |
lems | -0.577 |
Carney | -0.571 |
Token | , |
Token ID | 11 |
Feature activation | -3.13e-02 |
Pos logit contributions | |
ippers | +0.901 |
ipper | +0.880 |
rones | +0.698 |
lems | +0.694 |
Carney | +0.688 |
Neg logit contributions | |
LLOW | -1.003 |
tten | -0.980 |
�� | -0.961 |
Swordsman | -0.929 |
ADRA | -0.920 |
Token | culture |
Token ID | 3968 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
ippers | +0.023 |
ipper | +0.023 |
rones | +0.018 |
lems | +0.018 |
ploy | +0.018 |
Neg logit contributions | |
LLOW | -0.024 |
tten | -0.023 |
�� | -0.023 |
Swordsman | -0.022 |
FACE | -0.022 |
Token | , |
Token ID | 11 |
Feature activation | +5.18e-01 |
Pos logit contributions | |
ippers | +2.105 |
ipper | +2.061 |
rones | +1.641 |
lems | +1.633 |
Carney | +1.612 |
Neg logit contributions | |
LLOW | -2.376 |
tten | -2.328 |
�� | -2.285 |
Swordsman | -2.203 |
FACE | -2.185 |
Token | a |
Token ID | 257 |
Feature activation | +4.91e-01 |
Pos logit contributions | |
LLOW | +1.612 |
tten | +1.598 |
�� | +1.530 |
Swordsman | +1.509 |
FACE | +1.479 |
Neg logit contributions | |
ippers | -1.640 |
ipper | -1.585 |
rones | -1.281 |
lems | -1.279 |
ploy | -1.258 |
Token | very |
Token ID | 845 |
Feature activation | -7.98e-01 |
Pos logit contributions | |
LLOW | +0.374 |
tten | +0.370 |
�� | +0.355 |
Swordsman | +0.349 |
FACE | +0.342 |
Neg logit contributions | |
ippers | -0.367 |
ipper | -0.355 |
lems | -0.287 |
rones | -0.287 |
lihood | -0.281 |
Token | interesting |
Token ID | 3499 |
Feature activation | +9.91e-01 |
Pos logit contributions | |
ippers | +0.593 |
ipper | +0.576 |
rones | +0.465 |
lems | +0.460 |
lihood | +0.454 |
Neg logit contributions | |
LLOW | -0.614 |
tten | -0.602 |
�� | -0.583 |
Swordsman | -0.569 |
FACE | -0.562 |
Token | session |
Token ID | 6246 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
LLOW | +0.703 |
tten | +0.692 |
�� | +0.666 |
Swordsman | +0.650 |
FACE | +0.641 |
Neg logit contributions | |
ippers | -0.793 |
ipper | -0.769 |
rones | -0.634 |
lems | -0.631 |
ploy | -0.622 |
Token | to |
Token ID | 284 |
Feature activation | +8.38e-01 |
Pos logit contributions | |
ippers | +1.730 |
ipper | +1.694 |
rones | +1.339 |
Carney | +1.331 |
lems | +1.330 |
Neg logit contributions | |
LLOW | -1.928 |
tten | -1.879 |
�� | -1.849 |
Swordsman | -1.783 |
ADRA | -1.779 |
Token | say |
Token ID | 910 |
Feature activation | -3.66e-01 |
Pos logit contributions | |
LLOW | +0.671 |
tten | +0.660 |
�� | +0.629 |
Swordsman | +0.619 |
FACE | +0.613 |
Neg logit contributions | |
ippers | -0.593 |
ipper | -0.579 |
rones | -0.465 |
lems | -0.463 |
lihood | -0.460 |
Token | the |
Token ID | 262 |
Feature activation | +7.64e-01 |
Pos logit contributions | |
ippers | +0.275 |
ipper | +0.268 |
rones | +0.216 |
lems | +0.215 |
Carney | +0.212 |
Neg logit contributions | |
LLOW | -0.278 |
tten | -0.271 |
�� | -0.265 |
Swordsman | -0.257 |
FACE | -0.254 |
Token | least |
Token ID | 1551 |
Feature activation | -1.72e+00 |
Pos logit contributions | |
LLOW | +0.565 |
tten | +0.551 |
�� | +0.536 |
Swordsman | +0.517 |
FACE | +0.507 |
Neg logit contributions | |
ippers | -0.587 |
ipper | -0.571 |
rones | -0.466 |
lems | -0.462 |
lihood | -0.461 |
Token | . |
Token ID | 13 |
Feature activation | -9.00e-02 |
Pos logit contributions | |
ippers | +1.249 |
ipper | +1.222 |
rones | +0.974 |
lems | +0.969 |
Carney | +0.967 |
Neg logit contributions | |
LLOW | -1.341 |
tten | -1.309 |
�� | -1.287 |
Swordsman | -1.241 |
ADRA | -1.235 |
Token | I |
Token ID | 314 |
Feature activation | +3.33e-02 |
Pos logit contributions | |
ippers | +0.067 |
ipper | +0.066 |
rones | +0.052 |
lems | +0.052 |
ploy | +0.051 |
Neg logit contributions | |
LLOW | -0.069 |
tten | -0.067 |
�� | -0.066 |
Swordsman | -0.064 |
FACE | -0.063 |
Token | consider |
Token ID | 2074 |
Feature activation | +6.17e-01 |
Pos logit contributions | |
LLOW | +0.026 |
tten | +0.025 |
�� | +0.024 |
Swordsman | +0.024 |
FACE | +0.024 |
Neg logit contributions | |
ippers | -0.024 |
ipper | -0.024 |
lems | -0.019 |
rones | -0.019 |
lihood | -0.019 |
Token | Committee |
Token ID | 4606 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
LLOW | +0.697 |
tten | +0.669 |
�� | +0.658 |
Swordsman | +0.643 |
FACE | +0.627 |
Neg logit contributions | |
ippers | -0.757 |
ipper | -0.737 |
rones | -0.602 |
lems | -0.600 |
lihood | -0.591 |
Token | . |
Token ID | 13 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
ippers | +2.364 |
ipper | +2.304 |
Carney | +1.815 |
rones | +1.805 |
lems | +1.799 |
Neg logit contributions | |
LLOW | -2.769 |
tten | -2.698 |
�� | -2.647 |
Swordsman | -2.571 |
ADRA | -2.565 |
Token | F |
Token ID | 376 |
Feature activation | -3.65e-01 |
Pos logit contributions | |
LLOW | +1.241 |
tten | +1.193 |
�� | +1.169 |
Swordsman | +1.150 |
FACE | +1.130 |
Neg logit contributions | |
ippers | -1.232 |
ipper | -1.205 |
rones | -0.970 |
lems | -0.968 |
ploy | -0.950 |
Token | ess |
Token ID | 408 |
Feature activation | -4.59e-01 |
Pos logit contributions | |
ippers | +0.390 |
ipper | +0.384 |
rones | +0.332 |
lems | +0.331 |
ploy | +0.326 |
Neg logit contributions | |
LLOW | -0.163 |
tten | -0.154 |
�� | -0.152 |
Swordsman | -0.141 |
ADRA | -0.141 |
Token | end |
Token ID | 437 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
ippers | +0.397 |
ipper | +0.389 |
rones | +0.323 |
lems | +0.322 |
Carney | +0.320 |
Neg logit contributions | |
LLOW | -0.295 |
tten | -0.287 |
�� | -0.279 |
Swordsman | -0.268 |
FACE | -0.265 |
Token | en |
Token ID | 268 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
ippers | +1.780 |
ipper | +1.735 |
rones | +1.327 |
lems | +1.316 |
mast | +1.309 |
Neg logit contributions | |
LLOW | -2.417 |
�� | -2.345 |
tten | -2.314 |
ADRA | -2.291 |
Swordsman | -2.260 |
Token | took |
Token ID | 1718 |
Feature activation | -1.55e-01 |
Pos logit contributions | |
ippers | +1.037 |
ipper | +1.012 |
rones | +0.808 |
lems | +0.805 |
Carney | +0.797 |
Neg logit contributions | |
LLOW | -1.102 |
tten | -1.075 |
�� | -1.053 |
Swordsman | -1.019 |
ADRA | -1.007 |
Token | no |
Token ID | 645 |
Feature activation | -6.87e-01 |
Pos logit contributions | |
ippers | +0.115 |
ipper | +0.112 |
rones | +0.090 |
lems | +0.089 |
Carney | +0.089 |
Neg logit contributions | |
LLOW | -0.120 |
tten | -0.117 |
�� | -0.115 |
Swordsman | -0.111 |
ADRA | -0.110 |
Token | immediate |
Token ID | 7103 |
Feature activation | -2.21e+00 |
Pos logit contributions | |
ippers | +0.557 |
ipper | +0.543 |
rones | +0.444 |
lems | +0.442 |
lihood | +0.437 |
Neg logit contributions | |
LLOW | -0.482 |
tten | -0.468 |
�� | -0.456 |
Swordsman | -0.443 |
FACE | -0.437 |
Token | action |
Token ID | 2223 |
Feature activation | -1.86e+00 |
Pos logit contributions | |
ippers | +1.802 |
ipper | +1.768 |
rones | +1.450 |
lems | +1.446 |
Carney | +1.439 |
Neg logit contributions | |
LLOW | -1.524 |
tten | -1.484 |
�� | -1.452 |
Swordsman | -1.394 |
ADRA | -1.381 |
Token | , |
Token ID | 11 |
Feature activation | -3.10e-03 |
Pos logit contributions | |
ippers | +1.313 |
ipper | +1.285 |
lems | +1.013 |
rones | +1.011 |
Carney | +1.008 |
Neg logit contributions | |
LLOW | -1.500 |
tten | -1.458 |
�� | -1.436 |
ADRA | -1.384 |
Swordsman | -1.384 |
Token | . |
Token ID | 13 |
Feature activation | +3.60e-01 |
Pos logit contributions | |
ippers | +1.001 |
ipper | +0.979 |
rones | +0.788 |
lems | +0.785 |
Carney | +0.778 |
Neg logit contributions | |
LLOW | -0.996 |
tten | -0.971 |
�� | -0.950 |
Swordsman | -0.918 |
ADRA | -0.907 |
Token |
|
Token ID | 198 |
Feature activation | -6.69e-01 |
Pos logit contributions | |
LLOW | +0.280 |
tten | +0.271 |
�� | +0.265 |
Swordsman | +0.259 |
FACE | +0.255 |
Neg logit contributions | |
ippers | -0.265 |
ipper | -0.258 |
rones | -0.206 |
lems | -0.206 |
mast | -0.202 |
Token |
|
Token ID | 198 |
Feature activation | -7.59e-01 |
Pos logit contributions | |
ippers | +0.464 |
ipper | +0.452 |
rones | +0.355 |
lems | +0.353 |
Carney | +0.350 |
Neg logit contributions | |
LLOW | -0.548 |
tten | -0.534 |
�� | -0.520 |
Swordsman | -0.507 |
FACE | -0.504 |
Token | Impro |
Token ID | 23028 |
Feature activation | +1.14e+00 |
Pos logit contributions | |
ippers | +0.691 |
ipper | +0.678 |
rones | +0.568 |
lems | +0.567 |
Carney | +0.563 |
Neg logit contributions | |
LLOW | -0.457 |
tten | -0.441 |
�� | -0.429 |
Swordsman | -0.409 |
FACE | -0.408 |
Token | ving |
Token ID | 1075 |
Feature activation | +1.27e+00 |
Pos logit contributions | |
LLOW | +0.772 |
tten | +0.757 |
�� | +0.735 |
Swordsman | +0.703 |
FACE | +0.697 |
Neg logit contributions | |
ippers | -0.946 |
ipper | -0.926 |
rones | -0.761 |
lems | -0.760 |
Carney | -0.756 |
Token | delivery |
Token ID | 7585 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
LLOW | +0.963 |
tten | +0.942 |
�� | +0.916 |
Swordsman | +0.887 |
FACE | +0.880 |
Neg logit contributions | |
ippers | -0.943 |
ipper | -0.920 |
rones | -0.739 |
lems | -0.736 |
Carney | -0.724 |
Token | of |
Token ID | 286 |
Feature activation | -1.79e-01 |
Pos logit contributions | |
ippers | +1.132 |
ipper | +1.105 |
rones | +0.878 |
lems | +0.877 |
Carney | +0.866 |
Neg logit contributions | |
LLOW | -1.243 |
tten | -1.213 |
�� | -1.187 |
Swordsman | -1.150 |
FACE | -1.137 |
Token | services |
Token ID | 2594 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
ippers | +0.127 |
ipper | +0.124 |
rones | +0.098 |
lems | +0.098 |
Carney | +0.097 |
Neg logit contributions | |
LLOW | -0.143 |
tten | -0.139 |
�� | -0.136 |
Swordsman | -0.132 |
FACE | -0.130 |
Token | to |
Token ID | 284 |
Feature activation | -3.39e-01 |
Pos logit contributions | |
LLOW | +2.152 |
tten | +2.128 |
�� | +2.042 |
Swordsman | +2.020 |
FACE | +1.969 |
Neg logit contributions | |
ippers | -2.028 |
ipper | -1.975 |
rones | -1.588 |
lems | -1.582 |
lihood | -1.568 |
Token | the |
Token ID | 262 |
Feature activation | +4.63e-01 |
Pos logit contributions | |
ippers | +0.258 |
ipper | +0.254 |
rones | +0.205 |
lems | +0.203 |
Carney | +0.202 |
Neg logit contributions | |
LLOW | -0.253 |
tten | -0.248 |
�� | -0.240 |
Swordsman | -0.233 |
ADRA | -0.231 |
Token | stakeholders |
Token ID | 26138 |
Feature activation | +1.19e+00 |
Pos logit contributions | |
LLOW | +0.312 |
tten | +0.305 |
�� | +0.294 |
Swordsman | +0.283 |
FACE | +0.280 |
Neg logit contributions | |
ippers | -0.387 |
ipper | -0.380 |
rones | -0.316 |
lems | -0.311 |
Carney | -0.308 |
Token | � |
Token ID | 247 |
Feature activation | -2.53e-02 |
Pos logit contributions | |
LLOW | +0.328 |
�� | +0.320 |
tten | +0.319 |
ADRA | +0.300 |
Swordsman | +0.299 |
Neg logit contributions | |
ippers | -0.318 |
ipper | -0.310 |
rones | -0.252 |
lihood | -0.251 |
lems | -0.250 |
Token | t |
Token ID | 83 |
Feature activation | +1.26e+00 |
Pos logit contributions | |
ippers | +0.019 |
ipper | +0.018 |
rones | +0.015 |
lihood | +0.015 |
lems | +0.015 |
Neg logit contributions | |
tten | -0.019 |
LLOW | -0.019 |
�� | -0.018 |
FACE | -0.018 |
Swordsman | -0.018 |
Token | one |
Token ID | 530 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
LLOW | +0.964 |
tten | +0.939 |
�� | +0.915 |
Swordsman | +0.887 |
ADRA | +0.877 |
Neg logit contributions | |
ippers | -0.932 |
ipper | -0.911 |
rones | -0.734 |
lems | -0.733 |
mast | -0.720 |
Token | of |
Token ID | 286 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
LLOW | +1.360 |
tten | +1.329 |
�� | +1.284 |
Swordsman | +1.260 |
ADRA | +1.243 |
Neg logit contributions | |
ippers | -1.372 |
ipper | -1.334 |
lems | -1.096 |
rones | -1.085 |
lihood | -1.065 |
Token | the |
Token ID | 262 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
ippers | +2.443 |
ipper | +2.385 |
rones | +1.878 |
Carney | +1.878 |
lems | +1.865 |
Neg logit contributions | |
LLOW | -2.811 |
tten | -2.741 |
�� | -2.694 |
Swordsman | -2.615 |
ADRA | -2.570 |
Token | bureau |
Token ID | 10583 |
Feature activation | -4.10e+00 |
Pos logit contributions | |
ippers | +2.010 |
ipper | +1.963 |
rones | +1.581 |
lems | +1.574 |
Carney | +1.563 |
Neg logit contributions | |
LLOW | -1.989 |
tten | -1.939 |
�� | -1.899 |
Swordsman | -1.833 |
ADRA | -1.813 |
Token | � |
Token ID | 447 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
ippers | +2.900 |
ipper | +2.814 |
rones | +2.244 |
lems | +2.235 |
Carney | +2.186 |
Neg logit contributions | |
LLOW | -3.286 |
tten | -3.194 |
�� | -3.162 |
Swordsman | -3.057 |
FACE | -3.012 |
Token | � |
Token ID | 247 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
LLOW | +0.553 |
�� | +0.539 |
tten | +0.537 |
ADRA | +0.505 |
Swordsman | +0.504 |
Neg logit contributions | |
ippers | -0.527 |
ipper | -0.519 |
lihood | -0.417 |
rones | -0.415 |
lems | -0.412 |
Token | s |
Token ID | 82 |
Feature activation | -9.84e-01 |
Pos logit contributions | |
ippers | +0.647 |
ipper | +0.632 |
rones | +0.480 |
lems | +0.474 |
Carney | +0.474 |
Neg logit contributions | |
LLOW | -0.947 |
tten | -0.935 |
�� | -0.908 |
Swordsman | -0.883 |
FACE | -0.878 |
Token | many |
Token ID | 867 |
Feature activation | -1.26e+00 |
Pos logit contributions | |
ippers | +0.700 |
ipper | +0.686 |
rones | +0.544 |
lems | +0.541 |
Carney | +0.536 |
Neg logit contributions | |
LLOW | -0.784 |
tten | -0.766 |
�� | -0.748 |
Swordsman | -0.726 |
ADRA | -0.718 |
Token | tech |
Token ID | 7261 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
ippers | +0.908 |
ipper | +0.888 |
rones | +0.706 |
lems | +0.703 |
Carney | +0.695 |
Neg logit contributions | |
LLOW | -0.996 |
tten | -0.972 |
�� | -0.950 |
Swordsman | -0.922 |
ADRA | -0.912 |
Token | this |
Token ID | 428 |
Feature activation | +4.20e-01 |
Pos logit contributions | |
ippers | +0.369 |
ipper | +0.360 |
rones | +0.284 |
lems | +0.283 |
Carney | +0.279 |
Neg logit contributions | |
LLOW | -0.429 |
tten | -0.417 |
�� | -0.409 |
Swordsman | -0.397 |
FACE | -0.393 |
Token | all |
Token ID | 477 |
Feature activation | +2.35e+00 |
Pos logit contributions | |
LLOW | +0.325 |
tten | +0.317 |
�� | +0.308 |
Swordsman | +0.300 |
FACE | +0.296 |
Neg logit contributions | |
ippers | -0.310 |
ipper | -0.301 |
lems | -0.243 |
rones | -0.242 |
Carney | -0.237 |
Token | happens |
Token ID | 4325 |
Feature activation | +1.84e+00 |
Pos logit contributions | |
LLOW | +1.882 |
tten | +1.836 |
�� | +1.797 |
Swordsman | +1.739 |
FACE | +1.727 |
Neg logit contributions | |
ippers | -1.622 |
ipper | -1.588 |
rones | -1.258 |
lems | -1.251 |
lihood | -1.246 |
Token | from |
Token ID | 422 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
LLOW | +1.428 |
tten | +1.393 |
�� | +1.355 |
Swordsman | +1.330 |
FACE | +1.310 |
Neg logit contributions | |
ippers | -1.336 |
ipper | -1.300 |
Carney | -1.032 |
rones | -1.031 |
lems | -1.028 |
Token | Monday |
Token ID | 3321 |
Feature activation | +4.51e-01 |
Pos logit contributions | |
LLOW | +0.864 |
tten | +0.847 |
�� | +0.818 |
Swordsman | +0.800 |
FACE | +0.782 |
Neg logit contributions | |
ippers | -0.889 |
ipper | -0.867 |
lems | -0.699 |
rones | -0.697 |
lihood | -0.692 |
Token | April |
Token ID | 3035 |
Feature activation | -4.83e+00 |
Pos logit contributions | |
LLOW | +0.332 |
tten | +0.323 |
�� | +0.314 |
Swordsman | +0.305 |
FACE | +0.301 |
Neg logit contributions | |
ippers | -0.349 |
ipper | -0.340 |
rones | -0.275 |
lems | -0.274 |
lihood | -0.273 |
Token | 30 |
Token ID | 1542 |
Feature activation | -4.58e-01 |
Pos logit contributions | |
ippers | +3.522 |
ipper | +3.459 |
rones | +2.758 |
Carney | +2.758 |
lems | +2.748 |
Neg logit contributions | |
LLOW | -3.692 |
tten | -3.631 |
�� | -3.616 |
ADRA | -3.455 |
Swordsman | -3.404 |
Token | th |
Token ID | 400 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
ippers | +0.274 |
ipper | +0.266 |
rones | +0.200 |
lems | +0.199 |
Carney | +0.197 |
Neg logit contributions | |
LLOW | -0.418 |
tten | -0.409 |
�� | -0.401 |
Swordsman | -0.391 |
FACE | -0.387 |
Token | at |
Token ID | 379 |
Feature activation | +8.09e-01 |
Pos logit contributions | |
ippers | +0.761 |
ipper | +0.743 |
rones | +0.593 |
lems | +0.590 |
Carney | +0.584 |
Neg logit contributions | |
LLOW | -0.813 |
tten | -0.794 |
�� | -0.779 |
Swordsman | -0.752 |
ADRA | -0.745 |
Token | 1900 |
Token ID | 21489 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
LLOW | +0.588 |
tten | +0.574 |
�� | +0.552 |
Swordsman | +0.543 |
FACE | +0.531 |
Neg logit contributions | |
ippers | -0.636 |
ipper | -0.621 |
rones | -0.504 |
lihood | -0.502 |
lems | -0.500 |
Token | g |
Token ID | 308 |
Feature activation | -2.57e+00 |
Pos logit contributions | |
ippers | +0.993 |
ipper | +0.974 |
rones | +0.809 |
lems | +0.807 |
Carney | +0.801 |
Neg logit contributions | |
LLOW | -0.726 |
tten | -0.705 |
�� | -0.688 |
Swordsman | -0.660 |
ADRA | -0.652 |
Token | able |
Token ID | 1498 |
Feature activation | -1.40e+00 |
Pos logit contributions | |
ippers | +0.033 |
ipper | +0.032 |
rones | +0.026 |
lems | +0.025 |
mast | +0.025 |
Neg logit contributions | |
LLOW | -0.037 |
tten | -0.037 |
�� | -0.035 |
Swordsman | -0.034 |
FACE | -0.034 |
Token | to |
Token ID | 284 |
Feature activation | -6.13e-01 |
Pos logit contributions | |
ippers | +1.094 |
ipper | +1.073 |
lems | +0.867 |
rones | +0.867 |
Carney | +0.858 |
Neg logit contributions | |
LLOW | -1.021 |
tten | -0.986 |
�� | -0.972 |
ADRA | -0.934 |
Swordsman | -0.933 |
Token | prevent |
Token ID | 2948 |
Feature activation | -4.04e-01 |
Pos logit contributions | |
ippers | +0.451 |
ipper | +0.441 |
rones | +0.357 |
lems | +0.354 |
Carney | +0.348 |
Neg logit contributions | |
LLOW | -0.471 |
tten | -0.467 |
�� | -0.448 |
Swordsman | -0.438 |
FACE | -0.431 |
Token | the |
Token ID | 262 |
Feature activation | -1.89e-01 |
Pos logit contributions | |
ippers | +0.284 |
ipper | +0.277 |
rones | +0.220 |
lems | +0.218 |
mast | +0.214 |
Neg logit contributions | |
LLOW | -0.325 |
tten | -0.319 |
�� | -0.311 |
Swordsman | -0.302 |
FACE | -0.298 |
Token | Petro |
Token ID | 36863 |
Feature activation | +4.39e-01 |
Pos logit contributions | |
ippers | +0.129 |
ipper | +0.126 |
rones | +0.100 |
lems | +0.099 |
mast | +0.098 |
Neg logit contributions | |
LLOW | -0.155 |
tten | -0.151 |
�� | -0.148 |
Swordsman | -0.144 |
ADRA | -0.141 |
Token | car |
Token ID | 7718 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
LLOW | +0.427 |
tten | +0.422 |
�� | +0.410 |
Swordsman | +0.400 |
FACE | +0.400 |
Neg logit contributions | |
ippers | -0.233 |
ipper | -0.226 |
rones | -0.161 |
lihood | -0.161 |
lems | -0.160 |
Token | ibe |
Token ID | 32438 |
Feature activation | -1.82e+00 |
Pos logit contributions | |
ippers | +1.300 |
ipper | +1.232 |
rones | +0.626 |
lems | +0.623 |
Carney | +0.597 |
Neg logit contributions | |
LLOW | -4.952 |
tten | -4.858 |
�� | -4.813 |
Swordsman | -4.689 |
ADRA | -4.676 |
Token | deal |
Token ID | 1730 |
Feature activation | -7.02e-01 |
Pos logit contributions | |
ippers | +1.012 |
ipper | +0.980 |
rones | +0.719 |
lems | +0.715 |
Carney | +0.704 |
Neg logit contributions | |
LLOW | -1.742 |
tten | -1.711 |
�� | -1.679 |
Swordsman | -1.637 |
ADRA | -1.622 |
Token | for |
Token ID | 329 |
Feature activation | +8.86e-01 |
Pos logit contributions | |
ippers | +0.600 |
ipper | +0.588 |
rones | +0.487 |
lems | +0.486 |
Carney | +0.481 |
Neg logit contributions | |
LLOW | -0.460 |
tten | -0.447 |
�� | -0.435 |
Swordsman | -0.418 |
FACE | -0.413 |
Token | two |
Token ID | 734 |
Feature activation | +5.51e-01 |
Pos logit contributions | |
LLOW | +0.642 |
tten | +0.631 |
�� | +0.609 |
Swordsman | +0.590 |
FACE | +0.585 |
Neg logit contributions | |
ippers | -0.690 |
ipper | -0.676 |
rones | -0.553 |
lems | -0.547 |
mast | -0.542 |
Token | years |
Token ID | 812 |
Feature activation | +1.88e+00 |
Pos logit contributions | |
LLOW | +0.365 |
tten | +0.355 |
�� | +0.346 |
Swordsman | +0.333 |
ADRA | +0.329 |
Neg logit contributions | |
ippers | -0.466 |
ipper | -0.457 |
rones | -0.378 |
lems | -0.375 |
mast | -0.372 |
Token | executions |
Token ID | 30632 |
Feature activation | -2.65e+00 |
Pos logit contributions | |
LLOW | +0.621 |
tten | +0.608 |
�� | +0.592 |
Swordsman | +0.578 |
FACE | +0.567 |
Neg logit contributions | |
ippers | -0.613 |
ipper | -0.599 |
rones | -0.480 |
lems | -0.479 |
lihood | -0.472 |
Token | took |
Token ID | 1718 |
Feature activation | +2.23e+00 |
Pos logit contributions | |
ippers | +1.835 |
ipper | +1.786 |
rones | +1.407 |
lems | +1.388 |
Carney | +1.387 |
Neg logit contributions | |
LLOW | -2.162 |
tten | -2.110 |
�� | -2.079 |
Swordsman | -2.007 |
ADRA | -1.991 |
Token | place |
Token ID | 1295 |
Feature activation | +1.74e+00 |
Pos logit contributions | |
LLOW | +1.443 |
tten | +1.402 |
�� | +1.364 |
Swordsman | +1.313 |
FACE | +1.294 |
Neg logit contributions | |
ippers | -1.923 |
ipper | -1.887 |
rones | -1.564 |
lems | -1.562 |
Carney | -1.545 |
Token | on |
Token ID | 319 |
Feature activation | -2.02e-02 |
Pos logit contributions | |
LLOW | +1.350 |
tten | +1.317 |
�� | +1.285 |
Swordsman | +1.262 |
FACE | +1.235 |
Neg logit contributions | |
ippers | -1.267 |
ipper | -1.242 |
lems | -0.979 |
rones | -0.979 |
Carney | -0.964 |
Token | Sunday |
Token ID | 3502 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
ippers | +0.016 |
ipper | +0.016 |
rones | +0.013 |
lems | +0.013 |
Carney | +0.013 |
Neg logit contributions | |
LLOW | -0.014 |
tten | -0.014 |
�� | -0.014 |
Swordsman | -0.013 |
FACE | -0.013 |
Token | November |
Token ID | 3389 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
ippers | +1.046 |
ipper | +1.019 |
rones | +0.811 |
lems | +0.806 |
Carney | +0.801 |
Neg logit contributions | |
LLOW | -1.185 |
tten | -1.156 |
�� | -1.137 |
Swordsman | -1.094 |
ADRA | -1.093 |
Token | 3 |
Token ID | 513 |
Feature activation | -7.66e-01 |
Pos logit contributions | |
ippers | +3.069 |
ipper | +3.003 |
rones | +2.412 |
Carney | +2.404 |
lems | +2.398 |
Neg logit contributions | |
LLOW | -3.158 |
tten | -3.090 |
�� | -3.064 |
ADRA | -2.929 |
Swordsman | -2.901 |
Token | , |
Token ID | 11 |
Feature activation | +1.19e-01 |
Pos logit contributions | |
ippers | +0.535 |
ipper | +0.522 |
rones | +0.413 |
lems | +0.410 |
Carney | +0.406 |
Neg logit contributions | |
LLOW | -0.622 |
tten | -0.607 |
�� | -0.596 |
Swordsman | -0.576 |
ADRA | -0.572 |
Token | a |
Token ID | 257 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
LLOW | +0.092 |
tten | +0.090 |
�� | +0.088 |
Swordsman | +0.085 |
FACE | +0.084 |
Neg logit contributions | |
ippers | -0.087 |
ipper | -0.085 |
rones | -0.068 |
lems | -0.067 |
Carney | -0.067 |
Token | source |
Token ID | 2723 |
Feature activation | +1.80e-01 |
Pos logit contributions | |
ippers | +0.329 |
ipper | +0.322 |
rones | +0.269 |
lems | +0.269 |
Carney | +0.265 |
Neg logit contributions | |
LLOW | -0.231 |
tten | -0.224 |
�� | -0.218 |
Swordsman | -0.210 |
FACE | -0.207 |
Token | told |
Token ID | 1297 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
LLOW | +0.130 |
tten | +0.127 |
�� | +0.124 |
Swordsman | +0.120 |
FACE | +0.118 |
Neg logit contributions | |
ippers | -0.141 |
ipper | -0.137 |
rones | -0.112 |
lems | -0.112 |
Carney | -0.110 |
Token | of |
Token ID | 286 |
Feature activation | -2.93e-01 |
Pos logit contributions | |
ippers | +0.992 |
ipper | +0.968 |
rones | +0.767 |
lems | +0.765 |
Carney | +0.756 |
Neg logit contributions | |
LLOW | -1.109 |
tten | -1.083 |
�� | -1.061 |
Swordsman | -1.027 |
ADRA | -1.016 |
Token | his |
Token ID | 465 |
Feature activation | +9.32e-01 |
Pos logit contributions | |
ippers | +0.202 |
ipper | +0.196 |
lems | +0.157 |
rones | +0.156 |
ploy | +0.152 |
Neg logit contributions | |
LLOW | -0.239 |
tten | -0.235 |
�� | -0.228 |
Swordsman | -0.223 |
FACE | -0.221 |
Token | ret |
Token ID | 1005 |
Feature activation | +9.60e-01 |
Pos logit contributions | |
LLOW | +0.813 |
tten | +0.795 |
�� | +0.771 |
Swordsman | +0.760 |
FACE | +0.757 |
Neg logit contributions | |
ippers | -0.590 |
ipper | -0.567 |
lems | -0.442 |
rones | -0.437 |
ploy | -0.431 |
Token | raction |
Token ID | 7861 |
Feature activation | -3.07e-01 |
Pos logit contributions | |
LLOW | +0.757 |
tten | +0.748 |
�� | +0.722 |
Swordsman | +0.705 |
FACE | +0.694 |
Neg logit contributions | |
ippers | -0.675 |
ipper | -0.658 |
lems | -0.526 |
Carney | -0.523 |
rones | -0.522 |
Token | and |
Token ID | 290 |
Feature activation | +2.75e-01 |
Pos logit contributions | |
ippers | +0.233 |
ipper | +0.228 |
rones | +0.183 |
lems | +0.183 |
Carney | +0.180 |
Neg logit contributions | |
LLOW | -0.230 |
tten | -0.225 |
�� | -0.219 |
Swordsman | -0.213 |
FACE | -0.210 |
Token | silence |
Token ID | 9550 |
Feature activation | -4.12e+00 |
Pos logit contributions | |
LLOW | +0.212 |
tten | +0.207 |
�� | +0.202 |
Swordsman | +0.196 |
FACE | +0.194 |
Neg logit contributions | |
ippers | -0.203 |
ipper | -0.197 |
lems | -0.160 |
rones | -0.158 |
ploy | -0.155 |
Token | . |
Token ID | 13 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
ippers | +2.957 |
ipper | +2.883 |
rones | +2.307 |
Carney | +2.301 |
lems | +2.283 |
Neg logit contributions | |
LLOW | -3.243 |
tten | -3.175 |
�� | -3.105 |
ADRA | -3.026 |
Swordsman | -2.983 |
Token | This |
Token ID | 770 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
LLOW | +0.253 |
tten | +0.243 |
�� | +0.237 |
Swordsman | +0.236 |
FACE | +0.232 |
Neg logit contributions | |
ippers | -0.243 |
ipper | -0.237 |
lems | -0.189 |
rones | -0.188 |
ploy | -0.185 |
Token | year |
Token ID | 614 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
ippers | +0.363 |
ipper | +0.354 |
lems | +0.291 |
rones | +0.288 |
ploy | +0.282 |
Neg logit contributions | |
LLOW | -0.334 |
tten | -0.326 |
�� | -0.315 |
Swordsman | -0.308 |
FACE | -0.304 |
Token | , |
Token ID | 11 |
Feature activation | -7.19e-01 |
Pos logit contributions | |
LLOW | +1.140 |
tten | +1.112 |
�� | +1.080 |
Swordsman | +1.055 |
FACE | +1.039 |
Neg logit contributions | |
ippers | -1.136 |
ipper | -1.112 |
lems | -0.904 |
rones | -0.891 |
lihood | -0.878 |
Token | on |
Token ID | 319 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
ippers | +0.536 |
ipper | +0.523 |
lems | +0.420 |
rones | +0.420 |
lihood | +0.413 |
Neg logit contributions | |
LLOW | -0.550 |
tten | -0.535 |
�� | -0.522 |
Swordsman | -0.509 |
FACE | -0.501 |